2019-12-10 13:47:31 +00:00
|
|
|
%{
|
|
|
|
(* START HEADER *)
|
|
|
|
|
|
|
|
[@@@warning "-42"]
|
|
|
|
|
|
|
|
open Region
|
2019-12-12 14:35:07 +01:00
|
|
|
module AST = Parser_cameligo.AST
|
2019-12-10 13:47:31 +00:00
|
|
|
open AST
|
|
|
|
|
|
|
|
|
|
|
|
type 'a sequence_elements = {
|
|
|
|
s_elts : ('a, semi) Utils.nsepseq;
|
|
|
|
s_terminator : semi option
|
|
|
|
}
|
|
|
|
|
|
|
|
type 'a record_elements = {
|
|
|
|
r_elts : (field_assign reg, semi) Utils.nsepseq;
|
|
|
|
r_terminator : semi option
|
|
|
|
}
|
|
|
|
|
|
|
|
type 'a sequence_or_record =
|
|
|
|
PaSequence of 'a sequence_elements
|
|
|
|
| PaRecord of 'a record_elements
|
|
|
|
| PaSingleExpr of expr
|
|
|
|
|
|
|
|
(* END HEADER *)
|
|
|
|
%}
|
|
|
|
|
|
|
|
(* See [ParToken.mly] for the definition of tokens. *)
|
|
|
|
|
|
|
|
(* Entry points *)
|
|
|
|
|
|
|
|
%start contract interactive_expr
|
|
|
|
%type <AST.t> contract
|
|
|
|
%type <AST.expr> interactive_expr
|
|
|
|
|
|
|
|
|
|
|
|
%nonassoc Ident
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
%nonassoc COLON (* Solves a shift/reduce problem that happens with record
|
|
|
|
and sequences. To elaborate:
|
|
|
|
- sequence_or_record_in can be reduced to
|
|
|
|
expr -> Ident, but also to
|
|
|
|
field_assignment -> Ident.
|
2019-12-10 13:47:31 +00:00
|
|
|
*)
|
|
|
|
%%
|
|
|
|
|
|
|
|
(* RULES *)
|
|
|
|
|
|
|
|
(* The rule [sep_or_term(item,sep)] ("separated or terminated list")
|
|
|
|
parses a non-empty list of items separated by [sep], and optionally
|
|
|
|
terminated by [sep]. *)
|
|
|
|
|
|
|
|
sep_or_term_list(item,sep):
|
|
|
|
nsepseq(item,sep) {
|
|
|
|
$1, None
|
|
|
|
}
|
|
|
|
| nseq(item sep {$1,$2}) {
|
|
|
|
let (first,sep), tail = $1 in
|
|
|
|
let rec trans (seq, prev_sep as acc) = function
|
|
|
|
[] -> acc
|
|
|
|
| (item,next_sep)::others ->
|
|
|
|
trans ((prev_sep,item)::seq, next_sep) others in
|
|
|
|
let list, term = trans ([],sep) tail
|
|
|
|
in (first, List.rev list), Some term }
|
|
|
|
|
|
|
|
(* Compound constructs *)
|
|
|
|
|
|
|
|
par(X):
|
|
|
|
LPAR X RPAR {
|
|
|
|
let region = cover $1 $3
|
|
|
|
and value = {
|
|
|
|
lpar = $1;
|
|
|
|
inside = $2;
|
|
|
|
rpar = $3}
|
|
|
|
in {region; value}
|
|
|
|
}
|
|
|
|
|
|
|
|
braces(X):
|
|
|
|
LBRACE X RBRACE {
|
|
|
|
let region = cover $1 $3
|
|
|
|
and value = {
|
|
|
|
lpar = $1;
|
|
|
|
inside = $2;
|
|
|
|
rpar = $3}
|
|
|
|
in {region; value}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
(* Sequences
|
|
|
|
|
|
|
|
Series of instances of the same syntactical category have often to
|
|
|
|
be parsed, like lists of expressions, patterns etc. The simplest of
|
|
|
|
all is the possibly empty sequence (series), parsed below by
|
|
|
|
[seq]. The non-empty sequence is parsed by [nseq]. Note that the
|
|
|
|
latter returns a pair made of the first parsed item (the parameter
|
|
|
|
[X]) and the rest of the sequence (possibly empty). This way, the
|
|
|
|
OCaml typechecker can keep track of this information along the
|
|
|
|
static control-flow graph. The rule [sepseq] parses possibly empty
|
|
|
|
sequences of items separated by some token (e.g., a comma), and
|
|
|
|
rule [nsepseq] is for non-empty such sequences. See module [Utils]
|
|
|
|
for the types corresponding to the semantic actions of those
|
|
|
|
rules.
|
|
|
|
*)
|
|
|
|
|
|
|
|
(* Possibly empty sequence of items *)
|
|
|
|
|
|
|
|
seq(item):
|
|
|
|
(**) { [] }
|
|
|
|
| item seq(item) { $1::$2 }
|
|
|
|
|
|
|
|
(* Non-empty sequence of items *)
|
|
|
|
|
|
|
|
nseq(item):
|
|
|
|
item seq(item) { $1,$2 }
|
|
|
|
|
|
|
|
(* Non-empty separated sequence of items *)
|
|
|
|
|
|
|
|
nsepseq(item,sep):
|
|
|
|
item { $1, [] }
|
|
|
|
| item sep nsepseq(item,sep) { let h,t = $3 in $1, ($2,h)::t }
|
|
|
|
|
|
|
|
(* Possibly empy separated sequence of items *)
|
|
|
|
|
|
|
|
sepseq(item,sep):
|
|
|
|
(**) { None }
|
|
|
|
| nsepseq(item,sep) { Some $1 }
|
|
|
|
|
|
|
|
(* Helpers *)
|
|
|
|
|
|
|
|
%inline type_name : Ident { $1 }
|
|
|
|
%inline field_name : Ident { $1 }
|
|
|
|
%inline module_name : Constr { $1 }
|
|
|
|
%inline struct_name : Ident { $1 }
|
|
|
|
|
|
|
|
(* Non-empty comma-separated values (at least two values) *)
|
|
|
|
|
|
|
|
tuple(item):
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
item COMMA nsepseq(item,COMMA) {
|
2019-12-10 13:47:31 +00:00
|
|
|
let h,t = $3 in $1,($2,h)::t
|
|
|
|
}
|
|
|
|
|
|
|
|
(* Possibly empty semicolon-separated values between brackets *)
|
|
|
|
|
|
|
|
list(item):
|
|
|
|
LBRACKET sep_or_term_list(item, COMMA) RBRACKET {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let elements, terminator = $2 in
|
2019-12-10 13:47:31 +00:00
|
|
|
{ value =
|
|
|
|
{
|
|
|
|
compound = Brackets ($1,$3);
|
|
|
|
elements = Some elements;
|
|
|
|
terminator;
|
|
|
|
};
|
|
|
|
region = cover $1 $3
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
| LBRACKET RBRACKET {
|
|
|
|
let value = {
|
|
|
|
compound = Brackets ($1,$2);
|
|
|
|
elements = None;
|
|
|
|
terminator = None} in
|
|
|
|
let region = cover $1 $2
|
|
|
|
in {value; region}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
|
2019-12-10 13:47:31 +00:00
|
|
|
(* Main *)
|
|
|
|
|
|
|
|
contract:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
declarations EOF {
|
2019-12-10 13:47:31 +00:00
|
|
|
{decl = $1; eof=$2} }
|
|
|
|
|
|
|
|
declarations:
|
|
|
|
declaration { $1,[] : AST.declaration Utils.nseq }
|
|
|
|
| declaration declarations { Utils.nseq_cons $1 $2 }
|
|
|
|
|
|
|
|
declaration:
|
|
|
|
| type_decl SEMI { TypeDecl $1 }
|
|
|
|
| let_declaration SEMI { Let $1 }
|
|
|
|
|
|
|
|
(* Type declarations *)
|
|
|
|
|
|
|
|
type_decl:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
Type type_name EQ type_expr {
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover $1 (type_expr_to_region $4) in
|
|
|
|
let value = {
|
|
|
|
kwd_type = $1;
|
|
|
|
name = $2;
|
|
|
|
eq = $3;
|
|
|
|
type_expr = $4;
|
|
|
|
}
|
|
|
|
in {region; value}
|
|
|
|
}
|
|
|
|
|
|
|
|
type_expr:
|
|
|
|
cartesian { $1 }
|
|
|
|
| sum_type { TSum $1 }
|
|
|
|
| record_type { TRecord $1 }
|
|
|
|
|
|
|
|
cartesian:
|
|
|
|
fun_type COMMA nsepseq(fun_type,COMMA) {
|
|
|
|
let value = Utils.nsepseq_cons $1 $2 $3 in
|
|
|
|
let region = nsepseq_to_region type_expr_to_region value
|
|
|
|
in TProd {region; value}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
| fun_type { ($1 : type_expr) }
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
|
2019-12-10 13:47:31 +00:00
|
|
|
fun_type:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
core_type {
|
|
|
|
$1
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| core_type ARROW fun_type {
|
|
|
|
let region = cover (type_expr_to_region $1) (type_expr_to_region $3) in
|
2019-12-10 13:47:31 +00:00
|
|
|
TFun {region; value = ($1, $2, $3)}
|
|
|
|
}
|
|
|
|
|
|
|
|
core_type:
|
|
|
|
type_name {
|
|
|
|
TVar $1
|
|
|
|
}
|
|
|
|
| module_name DOT type_name {
|
|
|
|
let module_name = $1.value in
|
|
|
|
let type_name = $3.value in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let value = module_name ^ "." ^ type_name in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover $1.region $3.region
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in
|
2019-12-10 13:47:31 +00:00
|
|
|
TVar {region; value}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| type_constr LPAR nsepseq(core_type, COMMA) RPAR {
|
2019-12-10 13:47:31 +00:00
|
|
|
let arg_val = $3 in
|
|
|
|
let constr = $1 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let start = $1.region in
|
|
|
|
let stop = $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
let lpar, rpar = $2, $4 in
|
|
|
|
TApp Region.{value = constr, {
|
|
|
|
value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
lpar;
|
2019-12-10 13:47:31 +00:00
|
|
|
rpar;
|
|
|
|
inside = arg_val
|
|
|
|
};
|
|
|
|
region = cover lpar rpar;
|
|
|
|
}; region}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| par (type_expr) {
|
2019-12-10 13:47:31 +00:00
|
|
|
TPar $1
|
|
|
|
}
|
|
|
|
|
|
|
|
type_constr:
|
|
|
|
type_name { $1 }
|
|
|
|
|
|
|
|
sum_type:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
VBAR nsepseq(variant,VBAR) {
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = nsepseq_to_region (fun x -> x.region) $2
|
|
|
|
in {region; value = $2}
|
|
|
|
}
|
|
|
|
|
|
|
|
variant:
|
|
|
|
Constr LPAR cartesian RPAR {
|
|
|
|
let region = cover $1.region $4
|
|
|
|
and value = {constr = $1; arg = Some ($2, $3)}
|
|
|
|
in {region; value}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| Constr {
|
2019-12-10 13:47:31 +00:00
|
|
|
{region=$1.region; value= {constr=$1; arg=None}} }
|
|
|
|
|
|
|
|
record_type:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
LBRACE sep_or_term_list(field_decl,COMMA) RBRACE {
|
2019-12-10 13:47:31 +00:00
|
|
|
let ne_elements, terminator = $2 in
|
|
|
|
let region = cover $1 $3
|
|
|
|
and value = {
|
|
|
|
compound = Braces ($1,$3);
|
|
|
|
ne_elements;
|
|
|
|
terminator;
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in {region; value}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
type_expr_field:
|
|
|
|
core_type { $1 }
|
|
|
|
| sum_type { TSum $1 }
|
|
|
|
| record_type { TRecord $1 }
|
|
|
|
|
|
|
|
field_decl:
|
|
|
|
field_name {
|
|
|
|
let value = {field_name = $1; colon = Region.ghost; field_type = TVar $1}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in {region = $1.region; value}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
| field_name COLON type_expr_field {
|
|
|
|
let stop = type_expr_to_region $3 in
|
|
|
|
let region = cover $1.region stop
|
|
|
|
and value = {field_name = $1; colon = $2; field_type = $3}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in {region; value}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
(* Top-level non-recursive definitions *)
|
|
|
|
|
|
|
|
let_declaration:
|
|
|
|
Let let_binding {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let kwd_let = $1 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let binding, (region: Region.region) = $2 in
|
|
|
|
{value = kwd_let, binding; region}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
|
2019-12-10 13:47:31 +00:00
|
|
|
es6_func:
|
|
|
|
ARROW expr {
|
|
|
|
$1, $2
|
|
|
|
}
|
|
|
|
|
|
|
|
let_binding:
|
|
|
|
| Ident type_annotation? EQ expr {
|
|
|
|
let pattern = PVar $1 in
|
|
|
|
let start = pattern_to_region pattern in
|
|
|
|
let stop = expr_to_region $4 in
|
|
|
|
let region = cover start stop in
|
|
|
|
({binders = pattern, []; lhs_type=$2; eq=$3; let_rhs=$4}, region)
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| tuple(sub_irrefutable) type_annotation? EQ expr {
|
|
|
|
let h, t = $1 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = pattern_to_region h in
|
|
|
|
let stop = last (fun (region, _) -> region) t in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = cover start stop in
|
2019-12-10 13:47:31 +00:00
|
|
|
let pattern = PTuple { value = $1; region } in
|
|
|
|
let start = region in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
({binders = pattern, []; lhs_type=$2; eq=$3; let_rhs=$4}, region)
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| WILD type_annotation? EQ expr {
|
2019-12-10 13:47:31 +00:00
|
|
|
let pattern = PWild $1 in
|
|
|
|
let start = pattern_to_region pattern in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
({binders = pattern, []; lhs_type=$2; eq=$3; let_rhs=$4}, region)
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| unit type_annotation? EQ expr {
|
2019-12-10 13:47:31 +00:00
|
|
|
let pattern = PUnit $1 in
|
|
|
|
let start = pattern_to_region pattern in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
({binders = pattern, []; lhs_type=$2; eq=$3; let_rhs=$4}, region)
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| record_pattern type_annotation? EQ expr {
|
2019-12-10 13:47:31 +00:00
|
|
|
let pattern = PRecord $1 in
|
|
|
|
let start = pattern_to_region pattern in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
({binders = pattern, []; lhs_type=$2; eq=$3; let_rhs=$4}, region)
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| par(closed_irrefutable) type_annotation? EQ expr {
|
2019-12-10 13:47:31 +00:00
|
|
|
let pattern = PPar $1 in
|
|
|
|
let start = pattern_to_region pattern in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
({binders = pattern, []; lhs_type=$2; eq=$3; let_rhs=$4}, region)
|
|
|
|
}
|
|
|
|
|
|
|
|
type_annotation:
|
|
|
|
COLON type_expr { $1,$2 }
|
|
|
|
|
|
|
|
(* Patterns *)
|
|
|
|
|
|
|
|
irrefutable:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
tuple(sub_irrefutable) {
|
|
|
|
let h, t = $1 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = pattern_to_region h in
|
|
|
|
let stop = last (fun (region, _) -> region) t in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = cover start stop in
|
2019-12-10 13:47:31 +00:00
|
|
|
PTuple { value = $1; region }
|
|
|
|
}
|
|
|
|
| sub_irrefutable { $1 }
|
|
|
|
|
|
|
|
sub_irrefutable:
|
|
|
|
Ident { PVar $1 }
|
|
|
|
| WILD { PWild $1 }
|
|
|
|
| unit { PUnit $1 }
|
|
|
|
| record_pattern { PRecord $1 }
|
|
|
|
| par(closed_irrefutable) { PPar $1 }
|
|
|
|
|
|
|
|
closed_irrefutable:
|
|
|
|
irrefutable { $1 }
|
|
|
|
| constr_pattern { PConstr $1 }
|
|
|
|
| typed_pattern { PTyped $1 }
|
|
|
|
|
|
|
|
typed_pattern:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
irrefutable COLON type_expr {
|
|
|
|
let start = pattern_to_region $1 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let stop = type_expr_to_region $3 in
|
|
|
|
let region = cover start stop in
|
|
|
|
{
|
|
|
|
value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
pattern = $1;
|
|
|
|
colon = $2;
|
2019-12-10 13:47:31 +00:00
|
|
|
type_expr = $3
|
|
|
|
};
|
|
|
|
region
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
pattern:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
LBRACKET sub_pattern COMMA DOTDOTDOT sub_pattern RBRACKET {
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = pattern_to_region $2 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = pattern_to_region $5 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
let val_ = {value = $2, $3, $5; region} in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
PList (PCons val_)
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| tuple(sub_pattern) {
|
|
|
|
let h, t = $1 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = pattern_to_region h in
|
|
|
|
let stop = last (fun (region, _) -> region) t in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = cover start stop in
|
2019-12-10 13:47:31 +00:00
|
|
|
PTuple { value = $1; region }
|
|
|
|
}
|
|
|
|
| core_pattern { $1 }
|
|
|
|
|
|
|
|
sub_pattern:
|
|
|
|
par(sub_pattern) { PPar $1 }
|
|
|
|
| core_pattern { $1 }
|
|
|
|
|
|
|
|
core_pattern:
|
|
|
|
Ident { PVar $1 }
|
|
|
|
| WILD { PWild $1 }
|
|
|
|
| unit { PUnit $1 }
|
|
|
|
| Int { PInt $1 }
|
|
|
|
| True { PTrue $1 }
|
|
|
|
| False { PFalse $1 }
|
|
|
|
| Str { PString $1 }
|
|
|
|
| par(ptuple) { PPar $1 }
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| list(sub_pattern) { PList (PListComp $1) }
|
2019-12-10 13:47:31 +00:00
|
|
|
| constr_pattern { PConstr $1 }
|
|
|
|
| record_pattern { PRecord $1 }
|
|
|
|
|
|
|
|
record_pattern:
|
|
|
|
LBRACE sep_or_term_list(field_pattern,COMMA) RBRACE {
|
|
|
|
let ne_elements, terminator = $2 in
|
|
|
|
let region = cover $1 $3 in
|
|
|
|
let value = {
|
|
|
|
compound = Braces ($1,$3);
|
|
|
|
ne_elements;
|
|
|
|
terminator;
|
|
|
|
}
|
|
|
|
in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{region; value}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
field_pattern:
|
|
|
|
field_name EQ sub_pattern {
|
|
|
|
let start = $1.region in
|
|
|
|
let stop = pattern_to_region $3 in
|
|
|
|
let region = cover start stop in
|
|
|
|
{ value = {field_name=$1; eq=$2; pattern=$3}; region }
|
|
|
|
}
|
|
|
|
|
|
|
|
constr_pattern:
|
|
|
|
C_None { PNone $1 }
|
|
|
|
| C_Some sub_pattern {
|
|
|
|
let stop = pattern_to_region $2 in
|
|
|
|
let region = cover $1 stop
|
|
|
|
and value = $1, $2
|
|
|
|
in PSomeApp {value; region}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| Constr {
|
2019-12-10 13:47:31 +00:00
|
|
|
PConstrApp { value = $1, None; region = $1.region } }
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| Constr sub_pattern {
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover $1.region (pattern_to_region $2) in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
PConstrApp { value = $1, Some $2; region }
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
ptuple:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
tuple(sub_pattern) {
|
|
|
|
let h, t = $1 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = pattern_to_region h in
|
|
|
|
let stop = last (fun (region, _) -> region) t in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = cover start stop in
|
|
|
|
PTuple { value = $1; region }
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
unit:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
LPAR RPAR {
|
2019-12-10 13:47:31 +00:00
|
|
|
let the_unit = ghost, ghost in
|
|
|
|
let region = cover $1 $2 in
|
|
|
|
{ value = the_unit; region }
|
|
|
|
}
|
|
|
|
|
|
|
|
(* Expressions *)
|
|
|
|
|
|
|
|
interactive_expr:
|
|
|
|
expr EOF { $1 }
|
|
|
|
|
|
|
|
expr:
|
|
|
|
base_cond__open(expr) { $1 }
|
|
|
|
| switch_expr(base_cond) { ECase $1 }
|
|
|
|
|
|
|
|
|
|
|
|
base_cond__open(x):
|
|
|
|
base_expr(x)
|
|
|
|
| conditional(x) { $1 }
|
|
|
|
|
|
|
|
base_cond:
|
|
|
|
base_cond__open(base_cond) { $1 }
|
|
|
|
|
|
|
|
type_expr_simple_args:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
LPAR nsepseq(type_expr_simple, COMMA) RPAR {
|
2019-12-10 13:47:31 +00:00
|
|
|
$1, $2, $3
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
type_expr_simple:
|
|
|
|
core_expr_2 type_expr_simple_args? {
|
2019-12-10 13:47:31 +00:00
|
|
|
let args = $2 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let constr = match $1 with
|
2019-12-10 13:47:31 +00:00
|
|
|
| EVar i -> i
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| EProj {value = {struct_name; field_path; _}; region} ->
|
|
|
|
let path =
|
|
|
|
(Utils.nsepseq_foldl
|
|
|
|
(fun a e ->
|
|
|
|
match e with
|
2019-12-10 13:47:31 +00:00
|
|
|
| FieldName v -> a ^ "." ^ v.value
|
|
|
|
| Component {value = c, _; _} -> a ^ "." ^ c
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
)
|
2019-12-10 13:47:31 +00:00
|
|
|
struct_name.value
|
|
|
|
field_path
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
)
|
2019-12-10 13:47:31 +00:00
|
|
|
in
|
|
|
|
{value = path; region }
|
|
|
|
| EArith (Mutez {value = s, _; region })
|
|
|
|
| EArith (Int {value = s, _; region })
|
|
|
|
| EArith (Nat {value = s, _; region }) -> { value = s; region }
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| EString (String {value = s; region}) -> { value = s; region }
|
2019-12-10 13:47:31 +00:00
|
|
|
| ELogic (BoolExpr (True t)) -> { value = "true"; region = t }
|
|
|
|
| ELogic (BoolExpr (False f)) -> { value = "false"; region = f }
|
|
|
|
| _ -> failwith "Not supported"
|
|
|
|
in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
match args with
|
2019-12-10 13:47:31 +00:00
|
|
|
Some (lpar, args, rpar) -> (
|
|
|
|
let start = expr_to_region $1 in
|
|
|
|
let stop = rpar in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = cover start stop in
|
2019-12-10 13:47:31 +00:00
|
|
|
TApp {
|
|
|
|
value = constr, {
|
|
|
|
value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
inside = args;
|
|
|
|
lpar;
|
2019-12-10 13:47:31 +00:00
|
|
|
rpar
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
};
|
|
|
|
region};
|
2019-12-10 13:47:31 +00:00
|
|
|
region}
|
|
|
|
)
|
|
|
|
| None -> TVar constr
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| LPAR nsepseq(type_expr_simple, COMMA) RPAR {
|
2019-12-10 13:47:31 +00:00
|
|
|
TProd {value = $2; region = cover $1 $3}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| LPAR type_expr_simple ARROW type_expr_simple RPAR {
|
2019-12-10 13:47:31 +00:00
|
|
|
TFun {value = $2, $3, $4; region = cover $1 $5}
|
|
|
|
}
|
|
|
|
|
|
|
|
type_annotation_simple:
|
|
|
|
COLON type_expr_simple { $2 }
|
|
|
|
|
|
|
|
fun_expr:
|
|
|
|
disj_expr_level es6_func {
|
|
|
|
let arrow, body = $2 in
|
|
|
|
let kwd_fun = Region.ghost in
|
|
|
|
let start = expr_to_region $1 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region body in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
|
|
|
let rec arg_to_pattern = (function
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| EVar val_ -> PVar val_
|
2019-12-10 13:47:31 +00:00
|
|
|
| EAnnot {value = (EVar v, typ); region} ->
|
|
|
|
PTyped {value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
pattern = PVar v;
|
2019-12-10 13:47:31 +00:00
|
|
|
colon = Region.ghost;
|
|
|
|
type_expr = typ;
|
|
|
|
} ; region}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| EPar {value = {inside; lpar; rpar}; region} ->
|
2019-12-10 13:47:31 +00:00
|
|
|
PPar {value = {inside = arg_to_pattern inside; lpar; rpar}; region}
|
|
|
|
| EUnit u -> PUnit u
|
|
|
|
| _ -> failwith "Not supported"
|
|
|
|
)
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in
|
2019-12-10 13:47:31 +00:00
|
|
|
let fun_args_to_pattern = (function
|
|
|
|
| EAnnot {value = (ETuple {value = fun_args; _}, _); _} -> (* ((foo:x, bar) : type) *)
|
|
|
|
let bindings = List.map (fun arg -> arg_to_pattern (snd arg)) (snd fun_args) in
|
|
|
|
(arg_to_pattern (fst fun_args), bindings)
|
|
|
|
| EAnnot {value = (EPar {value = {inside = fun_arg ; _}; _}, _); _} -> (* ((foo:x, bar) : type) *)
|
|
|
|
(arg_to_pattern fun_arg, [])
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| EPar {value = {inside = fun_arg; _ }; _} ->
|
2019-12-10 13:47:31 +00:00
|
|
|
(arg_to_pattern fun_arg, [])
|
|
|
|
| EAnnot e -> (arg_to_pattern (EAnnot e), [])
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| ETuple {value = fun_args; _} ->
|
2019-12-10 13:47:31 +00:00
|
|
|
let bindings = List.map (fun arg -> arg_to_pattern (snd arg)) (snd fun_args) in
|
|
|
|
(arg_to_pattern (fst fun_args), bindings)
|
|
|
|
| EUnit e ->
|
|
|
|
(arg_to_pattern (EUnit e), [])
|
|
|
|
| _ -> failwith "Not supported"
|
|
|
|
)
|
|
|
|
in
|
|
|
|
let binders = fun_args_to_pattern $1 in
|
|
|
|
let f = {
|
|
|
|
kwd_fun ;
|
|
|
|
binders ;
|
|
|
|
lhs_type = None;
|
|
|
|
arrow ;
|
|
|
|
body ;
|
|
|
|
} in
|
|
|
|
EFun { region; value=f }
|
|
|
|
}
|
|
|
|
|
|
|
|
base_expr(right_expr):
|
|
|
|
let_expr(right_expr)
|
|
|
|
| disj_expr_level { $1 }
|
|
|
|
| fun_expr { $1 }
|
|
|
|
|
|
|
|
conditional(right_expr):
|
|
|
|
if_then_else(right_expr)
|
|
|
|
| if_then(right_expr) { ECond $1 }
|
|
|
|
|
|
|
|
parenthesized_expr:
|
|
|
|
braces (expr) { $1.value.inside }
|
|
|
|
| par (expr) { $1.value.inside }
|
|
|
|
|
|
|
|
if_then(right_expr):
|
|
|
|
If parenthesized_expr LBRACE closed_if RBRACE {
|
|
|
|
let the_unit = ghost, ghost in
|
|
|
|
let ifnot = EUnit {region=ghost; value=the_unit} in
|
|
|
|
let region = cover $1 $5 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{
|
2019-12-10 13:47:31 +00:00
|
|
|
value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
kwd_if = $1;
|
|
|
|
test = $2;
|
|
|
|
kwd_then = $3;
|
2019-12-10 13:47:31 +00:00
|
|
|
ifso = $4;
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
kwd_else = Region.ghost;
|
2019-12-10 13:47:31 +00:00
|
|
|
ifnot;
|
|
|
|
};
|
|
|
|
region
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if_then_else(right_expr):
|
|
|
|
If parenthesized_expr LBRACE closed_if SEMI RBRACE Else LBRACE right_expr SEMI RBRACE {
|
|
|
|
let region = cover $1 $11 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{
|
2019-12-10 13:47:31 +00:00
|
|
|
value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
kwd_if = $1;
|
|
|
|
test = $2;
|
|
|
|
kwd_then = $3;
|
2019-12-10 13:47:31 +00:00
|
|
|
ifso = $4;
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
kwd_else = $6;
|
2019-12-10 13:47:31 +00:00
|
|
|
ifnot = $9
|
|
|
|
};
|
|
|
|
region
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
base_if_then_else__open(x):
|
|
|
|
base_expr(x) { $1 }
|
|
|
|
| if_then_else(x) { ECond $1 }
|
|
|
|
|
|
|
|
base_if_then_else:
|
|
|
|
base_if_then_else__open(base_if_then_else) { $1 }
|
|
|
|
|
|
|
|
closed_if:
|
|
|
|
base_if_then_else__open(closed_if) { $1 }
|
|
|
|
| switch_expr(base_if_then_else) { ECase $1 }
|
|
|
|
|
|
|
|
switch_expr(right_expr):
|
|
|
|
Switch switch_expr_ LBRACE cases(right_expr) RBRACE {
|
|
|
|
let cases = $4 in
|
|
|
|
let start = $1 in
|
|
|
|
let stop = $5 in
|
|
|
|
let region = cover start stop in
|
|
|
|
{ value = {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
kwd_match = $1;
|
|
|
|
expr = $2;
|
|
|
|
lead_vbar = None;
|
2019-12-10 13:47:31 +00:00
|
|
|
kwd_with = Region.ghost;
|
|
|
|
cases = {
|
|
|
|
value = cases;
|
|
|
|
region = nsepseq_to_region (fun {region; _} -> region) $4
|
|
|
|
};
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
};
|
|
|
|
region
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
switch_expr_:
|
2019-12-10 13:47:31 +00:00
|
|
|
| par(expr) {
|
|
|
|
$1.value.inside
|
|
|
|
}
|
|
|
|
| core_expr_2 {
|
|
|
|
$1
|
|
|
|
}
|
|
|
|
|
|
|
|
cases(right_expr):
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
nseq(case_clause(right_expr)) {
|
2019-12-10 13:47:31 +00:00
|
|
|
let (hd, tl) = $1 in
|
|
|
|
hd, (List.map (fun f -> expr_to_region f.value.rhs, f) tl)
|
|
|
|
}
|
|
|
|
|
|
|
|
case_clause(right_expr):
|
|
|
|
VBAR pattern ARROW right_expr SEMI? {
|
|
|
|
let region = cover (pattern_to_region $2) (expr_to_region $4) in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{value =
|
2019-12-10 13:47:31 +00:00
|
|
|
{
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
pattern = $2;
|
|
|
|
arrow = $3;
|
|
|
|
rhs=$4
|
2019-12-10 13:47:31 +00:00
|
|
|
};
|
|
|
|
region
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
let_expr(right_expr):
|
|
|
|
Let let_binding SEMI right_expr {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let kwd_let = $1 in
|
|
|
|
let (binding: let_binding), _ = $2 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let kwd_in = $3 in
|
|
|
|
let body = $4 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = expr_to_region $4 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover $1 stop in
|
|
|
|
let let_in = {kwd_let; binding; kwd_in; body}
|
|
|
|
in ELetIn {region; value=let_in} }
|
|
|
|
|
|
|
|
disj_expr_level:
|
|
|
|
disj_expr { ELogic (BoolExpr (Or $1)) }
|
|
|
|
| conj_expr_level { $1 }
|
|
|
|
| par(tuple(disj_expr_level)) type_annotation_simple? {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = $1.region in
|
2019-12-10 13:47:31 +00:00
|
|
|
let tuple = ETuple {value=$1.value.inside; region} in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = match $2 with
|
2019-12-10 13:47:31 +00:00
|
|
|
| Some s -> cover $1.region (type_expr_to_region s)
|
|
|
|
| None -> region
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in
|
|
|
|
match $2 with
|
2019-12-10 13:47:31 +00:00
|
|
|
| Some typ -> EAnnot({value = tuple, typ; region})
|
|
|
|
| None -> tuple
|
|
|
|
}
|
|
|
|
|
|
|
|
bin_op(arg1,op,arg2):
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
arg1 op arg2 {
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = expr_to_region $1 in
|
|
|
|
let stop = expr_to_region $3 in
|
|
|
|
let region = cover start stop in
|
|
|
|
{ value = { arg1=$1; op=$2; arg2=$3}; region }
|
|
|
|
}
|
|
|
|
|
|
|
|
disj_expr:
|
|
|
|
bin_op(disj_expr_level, BOOL_OR, conj_expr_level)
|
|
|
|
| bin_op(disj_expr_level, Or, conj_expr_level) { $1 }
|
|
|
|
|
|
|
|
conj_expr_level:
|
|
|
|
conj_expr { ELogic (BoolExpr (And $1)) }
|
|
|
|
| comp_expr_level { $1 }
|
|
|
|
|
|
|
|
conj_expr:
|
|
|
|
bin_op(conj_expr_level, BOOL_AND, comp_expr_level) { $1 }
|
|
|
|
|
|
|
|
comp_expr_level:
|
|
|
|
lt_expr { ELogic (CompExpr (Lt $1)) }
|
|
|
|
| le_expr { ELogic (CompExpr (Leq $1)) }
|
|
|
|
| gt_expr { ELogic (CompExpr (Gt $1)) }
|
|
|
|
| ge_expr { ELogic (CompExpr (Geq $1)) }
|
|
|
|
| eq_expr { ELogic (CompExpr (Equal $1)) }
|
|
|
|
| ne_expr { ELogic (CompExpr (Neq $1)) }
|
|
|
|
| cat_expr_level { $1 }
|
|
|
|
|
|
|
|
lt_expr:
|
|
|
|
bin_op(comp_expr_level, LT, cat_expr_level) { $1 }
|
|
|
|
|
|
|
|
le_expr:
|
|
|
|
bin_op(comp_expr_level, LE, cat_expr_level) { $1 }
|
|
|
|
|
|
|
|
gt_expr:
|
|
|
|
bin_op(comp_expr_level, GT, cat_expr_level) { $1 }
|
|
|
|
|
|
|
|
ge_expr:
|
|
|
|
bin_op(comp_expr_level, GE, cat_expr_level) { $1 }
|
|
|
|
|
|
|
|
eq_expr:
|
|
|
|
bin_op(comp_expr_level, EQEQ, cat_expr_level) { $1 }
|
|
|
|
|
|
|
|
ne_expr:
|
|
|
|
bin_op(comp_expr_level, NE, cat_expr_level) { $1 }
|
|
|
|
|
|
|
|
cat_expr_level:
|
|
|
|
cat_expr { EString (Cat $1) }
|
|
|
|
| add_expr_level { $1 }
|
|
|
|
|
|
|
|
cat_expr:
|
|
|
|
bin_op(add_expr_level, CAT, add_expr_level) { $1 }
|
|
|
|
|
|
|
|
add_expr_level:
|
|
|
|
plus_expr { EArith (Add $1) }
|
|
|
|
| minus_expr { EArith (Sub $1) }
|
|
|
|
| mult_expr_level { $1 }
|
|
|
|
|
|
|
|
plus_expr:
|
|
|
|
bin_op(add_expr_level, PLUS, mult_expr_level) { $1 }
|
|
|
|
|
|
|
|
minus_expr:
|
|
|
|
bin_op(add_expr_level, MINUS, mult_expr_level) { $1 }
|
|
|
|
|
|
|
|
mult_expr_level:
|
|
|
|
times_expr { EArith (Mult $1) }
|
|
|
|
| div_expr { EArith (Div $1) }
|
|
|
|
| mod_expr { EArith (Mod $1) }
|
|
|
|
| unary_expr_level { $1 }
|
|
|
|
|
|
|
|
times_expr:
|
|
|
|
bin_op(mult_expr_level, TIMES, unary_expr_level) { $1 }
|
|
|
|
|
|
|
|
div_expr:
|
|
|
|
bin_op(mult_expr_level, SLASH, unary_expr_level) { $1 }
|
|
|
|
|
|
|
|
mod_expr:
|
|
|
|
bin_op(mult_expr_level, Mod, unary_expr_level) { $1 }
|
|
|
|
|
|
|
|
unary_expr_level:
|
|
|
|
MINUS call_expr_level {
|
|
|
|
let start = $1 in
|
|
|
|
let end_ = expr_to_region $2 in
|
|
|
|
let region = cover start end_
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
and value = {op = $1; arg = $2}
|
|
|
|
in EArith (Neg {region; value})
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
| NOT call_expr_level {
|
|
|
|
let start = $1 in
|
|
|
|
let end_ = expr_to_region $2 in
|
|
|
|
let region = cover start end_
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
and value = {op = $1; arg = $2} in
|
2019-12-10 13:47:31 +00:00
|
|
|
ELogic (BoolExpr (Not ({region; value})))
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
| call_expr_level {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
$1
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
call_expr_level:
|
2019-12-10 13:47:31 +00:00
|
|
|
call_expr_level_in type_annotation_simple? {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let region = match $2 with
|
2019-12-10 13:47:31 +00:00
|
|
|
| Some s -> cover (expr_to_region $1) (type_expr_to_region s)
|
|
|
|
| None -> expr_to_region $1
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in
|
2019-12-10 13:47:31 +00:00
|
|
|
match $2 with
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| Some t ->
|
|
|
|
EAnnot { value = $1, t; region }
|
2019-12-10 13:47:31 +00:00
|
|
|
| None -> $1
|
|
|
|
}
|
|
|
|
|
|
|
|
call_expr_level_in:
|
|
|
|
call_expr { $1 }
|
|
|
|
| constr_expr { $1 }
|
|
|
|
| core_expr { $1 }
|
|
|
|
|
|
|
|
constr_expr:
|
|
|
|
C_None {
|
|
|
|
EConstr (ENone $1)
|
|
|
|
}
|
|
|
|
| C_Some core_expr {
|
|
|
|
let region = cover $1 (expr_to_region $2)
|
|
|
|
in EConstr (ESomeApp {value = $1,$2; region})
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
|
|
|
| Constr core_expr? {
|
2019-12-10 13:47:31 +00:00
|
|
|
let start = $1.region in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let stop = match $2 with
|
2019-12-10 13:47:31 +00:00
|
|
|
| Some c -> expr_to_region c
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| None -> start
|
2019-12-10 13:47:31 +00:00
|
|
|
in
|
|
|
|
let region = cover start stop in
|
|
|
|
EConstr (EConstrApp { value = $1,$2; region})
|
|
|
|
}
|
|
|
|
|
|
|
|
call_expr:
|
|
|
|
core_expr LPAR nsepseq(expr, COMMA) RPAR {
|
|
|
|
let start = expr_to_region $1 in
|
|
|
|
let stop = $4 in
|
|
|
|
let region = cover start stop in
|
|
|
|
let hd, tl = $3 in
|
|
|
|
let tl = (List.map (fun (_, a) -> a) tl) in
|
|
|
|
ECall { value = $1, (hd, tl); region }
|
|
|
|
}
|
|
|
|
| core_expr unit {
|
|
|
|
let start = expr_to_region $1 in
|
|
|
|
let stop = $2.region in
|
|
|
|
let region = cover start stop in
|
|
|
|
ECall { value = $1, (EUnit $2, []); region }
|
|
|
|
}
|
|
|
|
|
|
|
|
core_expr_2:
|
|
|
|
Int { EArith (Int $1) }
|
|
|
|
| Mtz { EArith (Mutez $1) }
|
|
|
|
| Nat { EArith (Nat $1) }
|
|
|
|
| Ident | module_field { EVar $1 }
|
|
|
|
| projection { EProj $1 }
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| Str { EString (String $1) }
|
2019-12-10 13:47:31 +00:00
|
|
|
| unit { EUnit $1 }
|
|
|
|
| False { ELogic (BoolExpr (False $1)) }
|
|
|
|
| True { ELogic (BoolExpr (True $1)) }
|
|
|
|
| list(expr) { EList (EListComp $1) }
|
|
|
|
|
|
|
|
list_or_spread:
|
|
|
|
LBRACKET expr COMMA sep_or_term_list(expr, COMMA) RBRACKET {
|
|
|
|
let (e, terminator) = $4 in
|
|
|
|
let e = Utils.nsepseq_cons $2 $3 e in
|
|
|
|
EList (EListComp ({ value =
|
|
|
|
{
|
|
|
|
compound = Brackets ($1,$5);
|
|
|
|
elements = Some e;
|
|
|
|
terminator;
|
|
|
|
};
|
|
|
|
region = cover $1 $5
|
|
|
|
}))
|
|
|
|
}
|
|
|
|
| LBRACKET expr COMMA DOTDOTDOT expr RBRACKET {
|
|
|
|
let region = cover $1 $6 in
|
|
|
|
EList (ECons {value={arg1=$2; op=$4; arg2=$5}; region})
|
|
|
|
}
|
|
|
|
| LBRACKET expr RBRACKET {
|
|
|
|
EList (EListComp ({ value =
|
|
|
|
{
|
|
|
|
compound = Brackets ($1,$3);
|
|
|
|
elements = Some ($2, []);
|
|
|
|
terminator = None;
|
|
|
|
};
|
|
|
|
region = cover $1 $3
|
|
|
|
}))
|
|
|
|
}
|
|
|
|
| LBRACKET RBRACKET {
|
|
|
|
let value = {
|
|
|
|
compound = Brackets ($1,$2);
|
|
|
|
elements = None;
|
|
|
|
terminator = None} in
|
|
|
|
let region = cover $1 $2
|
|
|
|
in EList (EListComp ( {value; region}))
|
|
|
|
}
|
|
|
|
|
|
|
|
core_expr:
|
|
|
|
Int { EArith (Int $1) }
|
|
|
|
| Mtz { EArith (Mutez $1) }
|
|
|
|
| Nat { EArith (Nat $1) }
|
|
|
|
| Ident | module_field { EVar $1 }
|
|
|
|
| projection { EProj $1 }
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| Str { EString (String $1) }
|
2019-12-10 13:47:31 +00:00
|
|
|
| unit { EUnit $1 }
|
|
|
|
| False { ELogic (BoolExpr (False $1)) }
|
|
|
|
| True { ELogic (BoolExpr (True $1)) }
|
|
|
|
| list_or_spread { $1 }
|
|
|
|
| par(expr) { EPar $1 }
|
|
|
|
| sequence_or_record { $1 }
|
|
|
|
|
|
|
|
module_field:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
module_name DOT field_name {
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover $1.region $3.region in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{ value = $1.value ^ "." ^ $3.value; region }
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
selection:
|
|
|
|
| LBRACKET Int RBRACKET selection {
|
|
|
|
let r, (h, t) = $4 in
|
|
|
|
let result:((selection, dot) Utils.nsepseq) = (Component $2), (Region.ghost, h) :: t in
|
|
|
|
r, result
|
|
|
|
}
|
|
|
|
| DOT field_name selection {
|
|
|
|
let r, (h, t) = $3 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let result:((selection, dot) Utils.nsepseq) = (FieldName $2), ($1, h) :: t in
|
2019-12-10 13:47:31 +00:00
|
|
|
r, result
|
|
|
|
}
|
|
|
|
| DOT field_name {
|
|
|
|
$1, ((FieldName $2), [])
|
|
|
|
}
|
|
|
|
| LBRACKET Int RBRACKET {
|
|
|
|
Region.ghost, ((Component $2), [])
|
|
|
|
}
|
|
|
|
|
|
|
|
projection:
|
|
|
|
struct_name selection {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let start = $1.region in
|
|
|
|
let stop = nsepseq_to_region (function
|
|
|
|
| FieldName f -> f.region
|
2019-12-10 13:47:31 +00:00
|
|
|
| Component c -> c.region) (snd $2)
|
|
|
|
in
|
|
|
|
let region = cover start stop in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{ value =
|
2019-12-10 13:47:31 +00:00
|
|
|
{
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
struct_name = $1;
|
2019-12-10 13:47:31 +00:00
|
|
|
selector = fst $2;
|
|
|
|
field_path = snd $2
|
|
|
|
};
|
|
|
|
region
|
|
|
|
}
|
|
|
|
}
|
|
|
|
| module_name DOT field_name selection {
|
|
|
|
let module_name = $1 in
|
|
|
|
let field_name = $3 in
|
|
|
|
let value = module_name.value ^ "." ^ field_name.value in
|
|
|
|
let struct_name = {$1 with value} in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let start = $1.region in
|
|
|
|
let stop = nsepseq_to_region (function
|
|
|
|
| FieldName f -> f.region
|
2019-12-10 13:47:31 +00:00
|
|
|
| Component c -> c.region) (snd $4)
|
|
|
|
in
|
|
|
|
let region = cover start stop in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{ value =
|
2019-12-10 13:47:31 +00:00
|
|
|
{
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
struct_name;
|
|
|
|
selector = fst $4;
|
2019-12-10 13:47:31 +00:00
|
|
|
field_path = snd $4
|
|
|
|
};
|
|
|
|
region
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
sequence_or_record_in:
|
2019-12-10 13:47:31 +00:00
|
|
|
expr SEMI sep_or_term_list(expr,SEMI) {
|
|
|
|
let (e, _region) = $3 in
|
|
|
|
let e = Utils.nsepseq_cons $1 $2 e in
|
|
|
|
PaSequence { s_elts = e; s_terminator = None}
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| field_assignment COMMA sep_or_term_list(field_assignment,COMMA) {
|
2019-12-10 13:47:31 +00:00
|
|
|
let (e, _region) = $3 in
|
|
|
|
let e = Utils.nsepseq_cons $1 $2 e in
|
|
|
|
PaRecord { r_elts = e; r_terminator = None}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
| expr SEMI? {
|
|
|
|
PaSingleExpr $1
|
|
|
|
}
|
|
|
|
|
|
|
|
sequence_or_record:
|
|
|
|
LBRACE sequence_or_record_in RBRACE {
|
|
|
|
let compound = Braces($1, $3) in
|
|
|
|
let region = cover $1 $3 in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
match $2 with
|
2019-12-10 13:47:31 +00:00
|
|
|
| PaSequence s -> (
|
|
|
|
let value: expr injection = {
|
|
|
|
compound;
|
|
|
|
elements = Some s.s_elts;
|
|
|
|
terminator = s.s_terminator;
|
|
|
|
}
|
|
|
|
in
|
|
|
|
ESeq {value; region}
|
|
|
|
)
|
|
|
|
| PaRecord r -> (
|
|
|
|
let value: field_assign reg ne_injection = {
|
|
|
|
compound;
|
|
|
|
ne_elements = r.r_elts;
|
|
|
|
terminator = r.r_terminator;
|
|
|
|
}
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
in
|
2019-12-10 13:47:31 +00:00
|
|
|
ERecord {value; region}
|
|
|
|
)
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
| PaSingleExpr e -> e
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
field_assignment:
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
field_name {
|
|
|
|
{ value =
|
2019-12-10 13:47:31 +00:00
|
|
|
{
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
field_name = $1;
|
|
|
|
assignment = Region.ghost;
|
2019-12-10 13:47:31 +00:00
|
|
|
field_expr = EVar $1
|
|
|
|
};
|
|
|
|
region = $1.region
|
|
|
|
}
|
|
|
|
}
|
|
|
|
| field_name COLON expr {
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
let start = $1.region in
|
|
|
|
let stop = expr_to_region $3 in
|
2019-12-10 13:47:31 +00:00
|
|
|
let region = cover start stop in
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
{ value =
|
2019-12-10 13:47:31 +00:00
|
|
|
{
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
field_name = $1;
|
|
|
|
assignment = $2;
|
2019-12-10 13:47:31 +00:00
|
|
|
field_expr = $3
|
|
|
|
};
|
|
|
|
region
|
Refactoring of comments (for [dune build @doc]).
Refactoring of parsing command-line arguments
* The type [options] is now abstract and implemented as an
object type to avoid struggling with scoping and type
inference when record types share some common field names.
Refactoring of ParserLog for PascaLIGO and CameLIGO
* The immediate motivation behind that refactoring was to
remove the use of a couple of global references. A
consequence is that we have a nicer and more compact code, by
threading a state. The files [pascaligo/Tests/pp.ligo] and
[ligodity/Tests/pp.mligo].
* Another consequence is that the choice of making strings from
AST nodes depends on the CLI (offsets? mode?). After this
refactoring, that choice is hardcoded in the simplifiers in a
few places (TODO), waiting for a general solution that would
have all CL options flow through the compiler.
* I removed the use of vendors [x_option.ml], [x_map.ml] and
[x_list.ml] when handling optional values. (Less dependencies
this way.)
Refactoring of the ASTs
* I removed the node [local_decl], which was set to [[]]
already in a previous commit (which removed local
declarations as being redundant, as statements could already
be instructions or declarations).
* I changed [StrLit] to [String] in the AST of CameLIGO and
ReasonLIGO.
* I also changed the type [fun_expr] so now either a block is
present, and therefore followed by the [with] keyword, or it
is not. (Before, the presence of a block was not enforced in
the type with the presence of the keyword.)
Notes
* [LexerMain.ml] and [ParserMain.ml] for CameLIGO and PascaLIGO
are almost identical and differ in the same way (language
name and file extension), which suggests that they should be
in the [shared] folder and instanciated as a functor in the
future (TODO).
* I removed the blank characters at the end of many lines in
the parser of ReasonLIGO.
2019-12-13 12:21:52 +01:00
|
|
|
}
|
2019-12-10 13:47:31 +00:00
|
|
|
}
|