interfaces/Parsing_stats.atd

Rama main · réplica de lectura del repositorio original.

(*
   Output parsing stats for a collection of projects to json.
   This is used by semgrep-core -parsing_stats -json
*)

type t = {
  language: string; (* should be compatible with Lang.t *)
  global: project_stats;
  projects: project_stats list;
}

type project_stats = {
  name: string;

  (* 1 - error_line_count/line_count *)
  parsing_rate: float;

  (* Total number of lines that we attempted to parse. *)
  line_count: int;

  (* Number of lines that were unparseable.

     For a given file:
     - for tree-sitter parsers, this can be any value between 0 and the
       number of lines in the file.
     - for other parsers, it's either 0 or the total number of lines in
       the file.

     The number reported here is across all files that we attempted to parse.
     It's within [0, line_count].
  *)
  error_line_count: int;

  (* Number of files that we attempted to parse. *)
  file_count: int;

  (* Number of files that were parsed with one or more errors.
     It is in the range [0, file_count]. *)
  error_file_count: int;

  (*
     AST stats for evaluating the fraction of the AST that's made of
     nodes that need a proper translation. These numbers depend on which
     types of nodes counted, and of whether shallow todo nodes or
     raw subtrees are used in the AST. The latter will produce more nodes
     since they preserve whole sections of the original CST. Interpret
     with care.
  *)
  total_node_count: int;
  untranslated_node_count: int;
}