CLexer (coq-core.CLexer)

type starts_quotation =

| NoQuotation

| Quotation

When one registers a keyword she can declare it starts a quotation. In particular using QUOTATION("name:") in a grammar rule declares "name:" as a keyword and the token QUOTATION is matched whenever the keyword is followed by an identifier or a parenthesized text. Eg

constr:x string:.... ltac:(....) ltac:....

The delimiter is made of 1 or more occurrences of the same parenthesis, eg ((.....)) or [[[....]]]. The idea being that if the text happens to contain the closing delimiter, one can make the delimiter longer and avoid confusion (no escaping). Eg

string:[ .. ']' .. ]

Nesting the delimiter is allowed, eg ((..((...))..)) is OK.

Keywords don't need to end in ':'

val add_keyword : ?quotation:starts_quotation -> string -> unit

This should be functional but it is not due to the interface

val remove_keyword : string -> unit

val is_keyword : string -> bool

val keywords : unit -> CString.Set.t

type keyword_state

val set_keyword_state : keyword_state -> unit

val get_keyword_state : unit -> keyword_state

val check_ident : string -> unit

val is_ident : string -> bool

val check_keyword : string -> unit

val terminal : string -> string Tok.p

When string is not an ident, returns a keyword.

val terminal_number : string -> NumTok.Unsigned.t Tok.p

Precondition: the input is a number (c.f. NumTok.t)

val after : Loc.t -> Loc.t

after loc Will advance a lexing location as the lexer does; this can be used to implement parsing resumption from a given position:

let loc = Pcoq.Parsable.loc pa |> after in
let str = Gramlib.Stream.of_string text in
(* Stream.count being correct is critical for Coq's lexer *)
Gramlib.Stream.njunk loc.ep str;
let pa = Pcoq.Parsable.make ~loc str in
(* ready to resume parsing *)

The lexer of Coq:

module Lexer : Gramlib.Plexing.S with type te = Tok.t and type 'c pattern = 'c Tok.p

module Error : sig ... end

Create a lexer. true enables alternate handling for computing diffs. It ensures that, ignoring white space, the concatenated tokens equal the input string. Specifically:

for strings, return the enclosing quotes as tokens and treat the quoted value as if it was unquoted, possibly becoming multiple tokens
for comments, return the "(*" as a token and treat the contents of the comment as if it was not in a comment, possibly becoming multiple tokens
return any unrecognized Ascii or UTF-8 character as a string

module LexerDiff : Gramlib.Plexing.S with type te = Tok.t and type 'c pattern = 'c Tok.p