Skip to content
Snippets Groups Projects
Commit 21e86f5d authored by Wilke Pierre's avatar Wilke Pierre
Browse files

Mise à jour du squelette

parent 80c99974
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
.gitignore
+ 1
− 0
View file @ 21e86f5d
......@@ -20,6 +20,7 @@ alpaga/_build
grammar.html
src/generated_parser.ml
**/*.native
alpaga/ml_parser_generator
alpaga/alpaga
alpaga/ml_parser_generator.native
tykernel/*.ast
......
Sujet.pdf
+ 0
− 0
View file @ 21e86f5d
No preview for this file type
src/Makefile
+ 6
− 6
View file @ 21e86f5d
SRC=archi.ml ast.ml builtins.ml config.ml cfg_constprop.ml cfg_dead_assign.ml \
cfg.ml cfg_print.ml cfg_gen.ml cfg_nop_elim.ml cfg_run.ml elang.ml \
elang_print.ml elang_gen.ml elang_run.ml generated_parser.ml lexer_generator.ml \
linear_dse.ml linear_liveness.ml linear.ml linear_print.ml linear_gen.ml \
linear_run.ml ltl.ml ltl_print.ml ltl_gen.ml ltl_run.ml ltl_debug.ml main.ml \
options.ml parser.ml prog.ml regalloc.ml report.ml riscv.ml rtl.ml rtl_print.ml \
rtl_gen.ml rtl_run.ml symbols.ml utils.ml
elang_print.ml elang_gen.ml elang_run.ml e_regexp.ml generated_parser.ml \
lexer_generator.ml linear_dse.ml linear_liveness.ml linear.ml linear_print.ml \
linear_gen.ml linear_run.ml ltl.ml ltl_print.ml ltl_gen.ml ltl_run.ml \
ltl_debug.ml main.ml options.ml parser.ml prog.ml regalloc.ml report.ml \
riscv.ml rtl.ml rtl_print.ml rtl_gen.ml rtl_run.ml symbols.ml utils.ml
TG = main.native
all: $(TG)
$(TG): $(SRC)
ocamlbuild -cflags -warn-error,"+a" -use-ocamlfind $(TG)
ocamlbuild -cflags -warn-error,"+a-26" -use-ocamlfind $(TG)
clean:
rm -rf _build
......
src/e_regexp.ml 0 → 100644
+ 137
− 0
View file @ 21e86f5d
open Batteries
open Symbols
open Utils
(* Expressions régulières *)
(* Nous modélisons les expressions régulières avec le type suivant.
Une expressions régulière est soit :
- [Eps] qui dénote l'expressions vide.
- [Charset cs] dénote l'expression régulière qui matche l'ensemble
des caractères de l'ensemble [cs].
- [Cat(r1,r2)] dénote la concaténation de [r1] et [r2] : reconnaît les mots
[uv] tels que [u] appartient à [r1] et [v] appartient à [r2].
- [Alt(r1,r2)] dénote un choix entre [r1] et [r2] : reconnaît les mots reconnus
par [r1] et les mots reconnus par [r2].
- [Star r] dénote la répétition 0, 1 ou plusieurs fois de l'expression [r].
*)
type 'a set = 'a Set.t
type regexp =
| Eps
| Charset of char set
| Cat of regexp * regexp
| Alt of regexp * regexp
| Star of regexp
(* [char_regexp c] reconnaît le caractère [c] uniquement. *)
let char_regexp c = Charset (Set.singleton c)
(* [char_range l] reconnaît l'ensemble des caractères de [l]. *)
let char_range (l: char list) =
Charset (Set.of_list l)
(* [str_regexp s] reconnaît la chaîne de caractère [s]. *)
let str_regexp (s: char list) =
List.fold_right (fun c reg -> Cat(Charset (Set.singleton c), reg)) s Eps
(* [plus r] reconnaît 1 fois ou plus l'expression [r]. *)
let plus r = Cat(r,Star r)
(* Fonction d'affichage. Peut être utile pour déboguer. *)
let rec string_of_regexp r =
match r with
Eps -> "Eps"
| Charset c -> Printf.sprintf "[%s]" (string_of_char_list (Set.to_list c))
| Alt (r1,r2) -> Printf.sprintf "(%s)|(%s)"
(string_of_regexp r1) (string_of_regexp r2)
| Cat (r1,r2) -> Printf.sprintf "(%s)(%s)"
(string_of_regexp r1) (string_of_regexp r2)
| Star r -> Printf.sprintf "(%s)*" (string_of_regexp r)
(* La liste des expressions régulières permettant d'identifier les tokens du langage E *)
let list_regexp =
let lowercase_letters = "abcdefghijklmnopqrstuvwxyz" in
let uppercase_letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZ" in
let digits = "0123456789" in
let other_characters = "?!=<>_ ;,{}()[]-+*/%\n\t" in
let alphabet = char_list_of_string (lowercase_letters ^ uppercase_letters ^ digits ^ other_characters) in
let letter_regexp = char_range (char_list_of_string (uppercase_letters ^ lowercase_letters)) in
let digit_regexp = char_range (char_list_of_string digits) in
let keyword_regexp s = str_regexp (char_list_of_string s) in
[
(keyword_regexp "while", fun s -> Some (SYM_WHILE));
(keyword_regexp "int", fun s -> Some (SYM_INT));
(* begin TODO *)
(Eps, fun s -> Some (SYM_VOID));
(Eps, fun s -> Some (SYM_CHAR));
(Eps, fun s -> Some (SYM_IF));
(Eps, fun s -> Some (SYM_ELSE));
(Eps, fun s -> Some (SYM_RETURN));
(Eps, fun s -> Some (SYM_PRINT));
(Eps, fun s -> Some (SYM_STRUCT));
(Eps, fun s -> Some (SYM_POINT));
(Eps, fun s -> Some (SYM_PLUS));
(Eps, fun s -> Some (SYM_MINUS));
(Eps, fun s -> Some (SYM_ASTERISK));
(Eps, fun s -> Some (SYM_DIV));
(Eps, fun s -> Some (SYM_MOD));
(Eps, fun s -> Some (SYM_LBRACE));
(Eps, fun s -> Some (SYM_RBRACE));
(Eps, fun s -> Some (SYM_LBRACKET));
(Eps, fun s -> Some (SYM_RBRACKET));
(Eps, fun s -> Some (SYM_LPARENTHESIS));
(Eps, fun s -> Some (SYM_RPARENTHESIS));
(Eps, fun s -> Some (SYM_SEMICOLON));
(Eps, fun s -> Some (SYM_COMMA));
(Eps, fun s -> Some (SYM_ASSIGN));
(Eps, fun s -> Some (SYM_EQUALITY));
(Eps, fun s -> Some (SYM_NOTEQ));
(Eps, fun s -> Some (SYM_LT));
(Eps, fun s -> Some (SYM_GT));
(Eps, fun s -> Some (SYM_LEQ));
(Eps, fun s -> Some (SYM_GEQ));
(Eps, fun s -> Some (SYM_IDENTIFIER s));
(* end TODO *)
(Cat(keyword_regexp "//",
Cat(Star (char_range (List.filter (fun c -> c <> '\n') alphabet)),
Alt (char_regexp '\n', Eps))),
fun s -> None);
(Cat(keyword_regexp "/*",
Cat(
Star (Alt (char_range (List.filter (fun c -> c <> '*') alphabet),
Cat (char_regexp '*',
char_range (List.filter (fun c -> c <> '/') alphabet)))),
keyword_regexp "*/")),
fun s -> None);
(Cat (char_regexp '\'',
Cat (char_range (List.filter (fun c -> c <> '\'' && c <> '\\') alphabet),
char_regexp '\'')),
fun s -> Some (SYM_CHARACTER (String.get s 1)));
(Cat (char_regexp '\'', Cat (char_regexp '\\',
Cat (char_range (char_list_of_string "\\tn0"),
char_regexp '\''))),
fun s -> match String.get s 2 with
| '\\' -> Some (SYM_CHARACTER '\\')
| 'n' -> Some (SYM_CHARACTER '\n')
| 't' -> Some (SYM_CHARACTER '\t')
| '0' -> Some (SYM_CHARACTER '\x00')
| _ -> None
);
(Cat (char_regexp '"',
Cat (Star (
Alt (
char_range (List.filter (fun c -> c <> '"' && c <> '\\') alphabet),
Cat (char_regexp '\\', char_range (char_list_of_string "tn0\\\""))
)
),
char_regexp '"')),
fun s -> Some (SYM_STRING (Stdlib.Scanf.unescaped (String.slice ~first:1 ~last:(-1) s))));
(char_regexp ' ', fun s -> None);
(char_regexp '\n', fun s -> None);
(char_regexp '\t', fun s -> None);
(plus digit_regexp, fun s -> Some (SYM_INTEGER (int_of_string s)));
(Eps, fun s -> Some (SYM_EOF))
]
src/lexer_generator.ml
+ 38
− 181
View file @ 21e86f5d
......@@ -2,80 +2,10 @@ open Batteries
open BatList
open Symbols
open Utils
open E_regexp
type 'a set = 'a Set.t
let remove_dups l : 'a list =
List.fold_left (fun acc elt -> if List.mem elt acc then acc else elt::acc) [] l
let rec take n l =
if n = 0 then []
else match l with
| [] -> []
| a::r -> a::take (n-1) r
let char_list_of_string l : char list =
String.to_list l
let string_of_char_list cl =
String.of_list cl
let string_of_char_set s =
string_of_char_list (Set.to_list s)
let string_of_int_list l =
Printf.sprintf "%s" (String.concat "_" (List.map string_of_int l))
let string_of_int_set s =
string_of_int_list (Set.to_list s)
(* Expressions régulières *)
(* Nous modélisons les expressions régulières avec le type suivant.
Une expressions régulière est soit :
- [Eps] qui dénote l'expressions vide.
- [Charset cs] dénote l'expression régulière qui matche l'ensemble
des caractères de l'ensemble [cs].
- [Cat(r1,r2)] dénote la concaténation de [r1] et [r2] : reconnaît les mots
[uv] tels que [u] appartient à [r1] et [v] appartient à [r2].
- [Alt(r1,r2)] dénote un choix entre [r1] et [r2] : reconnaît les mots reconnus
par [r1] et les mots reconnus par [r2].
- [Star r] dénote la répétition 0, 1 ou plusieurs fois de l'expression [r].
*)
type regexp =
| Eps
| Charset of char set
| Cat of regexp * regexp
| Alt of regexp * regexp
| Star of regexp
(* [char_regexp c] reconnaît le caractère [c] uniquement. *)
let char_regexp c = Charset (Set.singleton c)
(* [char_range l] reconnaît l'ensemble des caractères de [l]. *)
let char_range (l: char list) =
Charset (Set.of_list l)
(* [str_regexp s] reconnaît la chaîne de caractère [s]. *)
let str_regexp (s: char list) =
List.fold_right (fun c reg -> Cat(Charset (Set.singleton c), reg)) s Eps
(* [plus r] reconnaît 1 fois ou plus l'expression [r]. *)
let plus r = Cat(r,Star r)
(* Fonction d'affichage. Peut être utile pour déboguer. *)
let rec string_of_regexp r =
match r with
Eps -> "Eps"
| Charset c -> Printf.sprintf "[%s]" (string_of_char_list (Set.to_list c))
| Alt (r1,r2) -> Printf.sprintf "(%s)|(%s)"
(string_of_regexp r1) (string_of_regexp r2)
| Cat (r1,r2) -> Printf.sprintf "(%s)(%s)"
(string_of_regexp r1) (string_of_regexp r2)
| Star r -> Printf.sprintf "(%s)*" (string_of_regexp r)
(* Non-deterministic Finite Automata (NFA) *)
(* Les états d'un NFA [nfa_state] sont des entiers.
......@@ -115,33 +45,40 @@ let empty_nfa =
(* Concaténation de NFAs. *)
let cat_nfa n1 n2 =
(* TODO *)
empty_nfa
(* Alternatives de NFAs *)
let alt_nfa n1 n2 =
(* TODO *)
empty_nfa
(* Répétition de NFAs *)
(* t est de type [string -> token option] *)
let star_nfa n t =
(* TODO *)
empty_nfa
(* [nfa_of_regexp r firststate t] construit un NFA qui reconnaît le même langage
(* [nfa_of_regexp r freshstate t] construit un NFA qui reconnaît le même langage
que l'expression régulière [r].
Pour ce faire, vous allez devoir générer des noms d'états pour vos automates.
Le paramètre [firststate]
[freshstate] correspond à un entier pour lequel il n'y a pas encore d'état dans
le nfa. Il suffit d'incrémenter [freshstate] pour obtenir de nouveaux état non utilisés.
[t] est une fonction du type [string -> token option] utile pour les état finaux.
*)
let rec nfa_of_regexp r firststate t =
let rec nfa_of_regexp r freshstate t =
match r with
| Eps -> { nfa_states = [firststate];
nfa_initial = [firststate];
nfa_final = [(firststate,t)];
nfa_step = fun q -> []}, firststate + 1
| Charset c -> { nfa_states = [firststate; firststate + 1];
nfa_initial = [firststate];
nfa_final = [firststate + 1, t];
nfa_step = fun q -> if q = firststate then [(Some c, firststate + 1)] else []
}, firststate + 2
| _ -> empty_nfa, firststate
| Eps -> { nfa_states = [freshstate];
nfa_initial = [freshstate];
nfa_final = [(freshstate,t)];
nfa_step = fun q -> []}, freshstate + 1
| Charset c -> { nfa_states = [freshstate; freshstate + 1];
nfa_initial = [freshstate];
nfa_final = [freshstate + 1, t];
nfa_step = fun q -> if q = freshstate then [(Some c, freshstate + 1)] else []
}, freshstate + 2
(* TODO *)
| _ -> empty_nfa, freshstate
(* Deterministic Finite Automaton (DFA) *)
......@@ -182,6 +119,7 @@ let epsilon_closure (n: nfa) (s: nfa_state) : nfa_state set =
(* La fonction [traversal visited s] effectue un parcours de l'automate en
partant de l'état [s], et en suivant uniquement les epsilon-transitions. *)
let rec traversal (visited: nfa_state set) (s: nfa_state) : nfa_state set =
(* TODO *)
visited
in
traversal Set.empty s
......@@ -189,10 +127,12 @@ let epsilon_closure (n: nfa) (s: nfa_state) : nfa_state set =
(* [epsilon_closure_set n ls] calcule l'union des epsilon-fermeture de chacun
des états du NFA [n] dans l'ensemble [ls]. *)
let epsilon_closure_set (n: nfa) (ls: nfa_state set) : nfa_state set =
(* TODO *)
ls
(* [dfa_initial_state n] calcule l'état initial de l'automate déterminisé. *)
let dfa_initial_state (n: nfa) : dfa_state =
(* TODO *)
Set.empty
(* Construction de la table de transitions de l'automate DFA. *)
......@@ -212,7 +152,7 @@ let dfa_initial_state (n: nfa) : dfa_state =
[(c1,q1);(c1,q2);...;(c1,qn);(c2,q'1);...(c2,q'm)] ->
[(c1,{q1,q2,...,qn});(c2,{q'1,...,q'm})] : [assoc_merge_vals]
- on applique la epsilon-fermeture sur tous les états:
[(c1,q1);...;(cn,qn)] -> [(c1, eps(q1)); ...; (cn, eps(qn))] :
[(c1,{q1,q2,...,qn});...;(cn,{qn}])] -> [(c1, eps({q1,q2,...,qn})); ...; (cn, eps({qn}))] :
[epsilon_closure_set]
On obtient alors l'ensemble des transitions depuis l'état [q] dans
......@@ -228,6 +168,11 @@ let assoc_throw_none (l : ('a option * 'b) list) : ('a * 'b) list =
| Some x -> Some (x,n)
) l
let assoc_distribute_key (l : ('a set * 'b) list) : ('a * 'b) list =
List.fold_left (fun (acc : ('a * 'b) list) (k, v) ->
Set.fold (fun c acc -> (c, v)::acc) k acc)
[] l
let assoc_merge_vals (l : ('a * 'b) list) : ('a * 'b set) list =
List.fold_left (fun (acc : ('a * 'b set) list) (k, v) ->
match List.assoc_opt k acc with
......@@ -235,19 +180,16 @@ let assoc_merge_vals (l : ('a * 'b) list) : ('a * 'b set) list =
| Some vl -> (k, Set.add v vl)::List.remove_assoc k acc
) [] l
let assoc_distribute_key (l : ('a set * 'b) list) : ('a * 'b) list =
List.fold_left (fun (acc : ('a * 'b) list) (k, v) ->
Set.fold (fun c acc -> (c, v)::acc) k acc)
[] l
let rec build_dfa_table (table: (dfa_state, (char * dfa_state) list) Hashtbl.t)
(n: nfa)
(ds: dfa_state) : unit =
match Hashtbl.find_option table ds with
| Some _ -> ()
| None ->
(* [transitions] contient les transitions du DFA construites
* à partir des transitions du NFA comme décrit auparavant *)
let transitions : (char * dfa_state) list =
(* TODO *)
[]
in
Hashtbl.replace table ds transitions;
......@@ -283,12 +225,14 @@ let priority t =
(* [min_priority l] renvoie le token de [l] qui a la plus petite priorité, ou
[None] si la liste [l] est vide. *)
let min_priority (l: token list) : token option =
(* TODO *)
None
(* [dfa_final_states n dfa_states] renvoie la liste des états finaux du DFA,
accompagnés du token qu'ils reconnaissent. *)
let dfa_final_states (n: nfa) (dfa_states: dfa_state list) :
(dfa_state * (string -> token option)) list =
(* TODO *)
[]
(* Construction de la relation de transition du DFA. *)
......@@ -297,6 +241,7 @@ let dfa_final_states (n: nfa) (dfa_states: dfa_state list) :
est la table générée par [build_dfa_table], définie ci-dessus. *)
let make_dfa_step (table: (dfa_state, (char * dfa_state) list) Hashtbl.t) =
fun (q: dfa_state) (a: char) ->
(* TODO *)
None
(* Finalement, on assemble tous ces morceaux pour construire l'automate. La
......@@ -473,99 +418,11 @@ let dfa_to_dot oc (n : dfa) (cl: char list): unit =
) n.dfa_states;
Printf.fprintf oc "}\n"
let alts l =
match l with
[] -> Eps
| a::r -> List.fold_left (fun acc r -> Alt(acc,r)) a l
let list_regexp =
let lowercase_letters = "abcdefghijklmnopqrstuvwxyz" in
let uppercase_letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZ" in
let digits = "0123456789" in
let other_characters = "?!=<>_ ;,{}()[]-+*/%\n\t" in
let alphabet = char_list_of_string (lowercase_letters ^ uppercase_letters ^ digits ^ other_characters) in
let letter_regexp = char_range (char_list_of_string (uppercase_letters ^ lowercase_letters)) in
let digit_regexp = char_range (char_list_of_string digits) in
let keyword_regexp s = str_regexp (char_list_of_string s) in
[
(keyword_regexp "while", fun s -> Some (SYM_WHILE));
(keyword_regexp "int", fun s -> Some (SYM_INT));
(keyword_regexp "void", fun s -> Some (SYM_VOID));
(keyword_regexp "char", fun s -> Some (SYM_CHAR));
(keyword_regexp "if", fun s -> Some (SYM_IF));
(keyword_regexp "else", fun s -> Some (SYM_ELSE));
(keyword_regexp "return", fun s -> Some (SYM_RETURN));
(keyword_regexp "print", fun s -> Some (SYM_PRINT));
(keyword_regexp "struct", fun s -> Some (SYM_STRUCT));
(keyword_regexp ".", fun s -> Some (SYM_POINT));
(keyword_regexp "+", fun s -> Some (SYM_PLUS));
(keyword_regexp "-", fun s -> Some (SYM_MINUS));
(keyword_regexp "*", fun s -> Some (SYM_ASTERISK));
(keyword_regexp "/", fun s -> Some (SYM_DIV));
(keyword_regexp "%", fun s -> Some (SYM_MOD));
(keyword_regexp "{", fun s -> Some (SYM_LBRACE));
(keyword_regexp "}", fun s -> Some (SYM_RBRACE));
(keyword_regexp "[", fun s -> Some (SYM_LBRACKET));
(keyword_regexp "]", fun s -> Some (SYM_RBRACKET));
(keyword_regexp "(", fun s -> Some (SYM_LPARENTHESIS));
(keyword_regexp ")", fun s -> Some (SYM_RPARENTHESIS));
(keyword_regexp ";", fun s -> Some (SYM_SEMICOLON));
(keyword_regexp ",", fun s -> Some (SYM_COMMA));
(keyword_regexp "=", fun s -> Some (SYM_ASSIGN));
(keyword_regexp "==", fun s -> Some (SYM_EQUALITY));
(keyword_regexp "!=", fun s -> Some (SYM_NOTEQ));
(keyword_regexp "<", fun s -> Some (SYM_LT));
(keyword_regexp ">", fun s -> Some (SYM_GT));
(keyword_regexp "<=", fun s -> Some (SYM_LEQ));
(keyword_regexp ">=", fun s -> Some (SYM_GEQ));
(Cat(letter_regexp,
Star(Alt(letter_regexp,
Alt(digit_regexp,
Charset (Set.singleton '_'))))),
fun s -> Some (SYM_IDENTIFIER s));
(Cat(keyword_regexp "//",
Cat(Star (char_range (List.filter (fun c -> c <> '\n') alphabet)),
Alt (char_regexp '\n', Eps))),
fun s -> None);
(Cat(keyword_regexp "/*",
Cat(
Star (Alt (char_range (List.filter (fun c -> c <> '*') alphabet),
Cat (char_regexp '*',
char_range (List.filter (fun c -> c <> '/') alphabet)))),
keyword_regexp "*/")),
fun s -> None);
(Cat (char_regexp '\'',
Cat (char_range (List.filter (fun c -> c <> '\'' && c <> '\\') alphabet),
char_regexp '\'')),
fun s -> Some (SYM_CHARACTER (String.get s 1)));
(Cat (char_regexp '\'', Cat (char_regexp '\\',
Cat (char_range (char_list_of_string "\\tn0"),
char_regexp '\''))),
fun s -> match String.get s 2 with
| '\\' -> Some (SYM_CHARACTER '\\')
| 'n' -> Some (SYM_CHARACTER '\n')
| 't' -> Some (SYM_CHARACTER '\t')
| '0' -> Some (SYM_CHARACTER '\x00')
| _ -> None
);
(Cat (char_regexp '"',
Cat (Star (
Alt (
char_range (List.filter (fun c -> c <> '"' && c <> '\\') alphabet),
Cat (char_regexp '\\', char_range (char_list_of_string "tn0\\\""))
)
),
char_regexp '"')),
fun s -> Some (SYM_STRING (Stdlib.Scanf.unescaped (String.slice ~first:1 ~last:(-1) s))));
(char_regexp ' ', fun s -> None);
(char_regexp '\n', fun s -> None);
(char_regexp '\t', fun s -> None);
(plus digit_regexp, fun s -> Some (SYM_INTEGER (int_of_string s)));
(Eps, fun s -> Some (SYM_EOF))
]
let nfa_of_list_regexp l =
let (n, fs) = List.fold_left (fun (nfa, fs) (r,t) ->
let n,fs = nfa_of_regexp r fs t in
......
src/utils.ml
+ 26
− 0
View file @ 21e86f5d
......@@ -251,3 +251,29 @@ let assoc_err ?word:(word="item") k l =
| Some v -> OK v
| None -> Error (Format.sprintf "%s %s not found." word k)
let remove_dups l : 'a list =
List.fold_left (fun acc elt -> if List.mem elt acc then acc else elt::acc) [] l
let rec take n l =
if n = 0 then []
else match l with
| [] -> []
| a::r -> a::take (n-1) r
let char_list_of_string l : char list =
String.to_list l
let string_of_char_list cl =
String.of_list cl
let string_of_char_set s =
string_of_char_list (Set.to_list s)
let string_of_int_list l =
Printf.sprintf "%s" (String.concat "_" (List.map string_of_int l))
let string_of_int_set s =
string_of_int_list (Set.to_list s)
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment