wip explain matcher

lib/matchers/languages.ml

@@ -1,4 +1,4 @@
-let enable_alphanum_delimiters = true
+open Syntax_config
 
 module Text = struct
   module Syntax = struct
@@ -156,21 +156,19 @@ module Ruby = struct
 
     let user_defined_delimiters =
       Generic.Syntax.user_defined_delimiters
-      @ if enable_alphanum_delimiters then
-        [ "class", "end"
-        ; "def", "end"
-        ; "do", "end"
-        ; "if", "end"
-        ; "case", "end"
-        ; "unless", "end"
-        ; "while", "end"
-        ; "until", "end"
-        ; "for", "end"
-        ; "begin", "end"
-        ; "module", "end"
-        ]
-      else
-        []
+      @
+      [ "class", "end"
+      ; "def", "end"
+      ; "do", "end"
+      ; "if", "end"
+      ; "case", "end"
+      ; "unless", "end"
+      ; "while", "end"
+      ; "until", "end"
+      ; "for", "end"
+      ; "begin", "end"
+      ; "module", "end"
+      ]
 
     let raw_string_literals =
       [ ({|"|}, {|"|})
@@ -192,16 +190,14 @@ module Elixir = struct
 
     let user_defined_delimiters =
       Generic.Syntax.user_defined_delimiters
-      @ if enable_alphanum_delimiters then
-        [ "fn", "end"
-        ; "do", "end"
-        ; "case", "end"
-        ; "cond", "end"
-        ; "if", "end"
-        ; "<", ">"
-        ]
-      else
-        []
+      @
+      [ "fn", "end"
+      ; "do", "end"
+      ; "case", "end"
+      ; "cond", "end"
+      ; "if", "end"
+      ; "<", ">"
+      ]
 
     let raw_string_literals =
       [ ({|"""|}, {|"""|})
@@ -268,13 +264,11 @@ module Erlang = struct
 
     let user_defined_delimiters =
       Generic.Syntax.user_defined_delimiters
-      @ if enable_alphanum_delimiters then
-        [ "fun", "end"
-        ; "case", "end"
-        ; "if", "end"
-        ]
-      else
-        []
+      @
+      [ "fun", "end"
+      ; "case", "end"
+      ; "if", "end"
+      ]
 
     let comment_parser =
       [ Until_newline "%"
@@ -388,13 +382,11 @@ module OCaml = struct
 
     let user_defined_delimiters =
       Generic.Syntax.user_defined_delimiters
-      @ if enable_alphanum_delimiters then
-        [ "begin", "end"
-        ; "struct", "end"
-        ; "sig", "end"
-        ]
-      else
-        []
+      @
+      [ "begin", "end"
+      ; "struct", "end"
+      ; "sig", "end"
+      ]
 
 
     (* Override ' as escapable string literal, since
@@ -531,16 +523,19 @@ let select_with_extension extension : (module Types.Matcher.S) =
   | ".txt" -> (module Text)
   | _ -> (module Generic)
 
-let create (t : Syntax_config.t) =
+let create
+    { user_defined_delimiters
+    ; escapable_string_literals
+    ; escape_char
+    ; raw_string_literals
+    ; comment_parser
+    } =
   let module User_language = struct
-    let user_defined_delimiters = t.user_defined_delimiters
-
-    let escapable_string_literals = t.escapable_string_literals
-
-    let escape_char = t.escape_char
-
-    let raw_string_literals = t.raw_string_literals
-
-    let comment_parser = t.comment_parser
-  end in
+    let user_defined_delimiters = user_defined_delimiters
+    let escapable_string_literals = escapable_string_literals
+    let escape_char = escape_char
+    let raw_string_literals = raw_string_literals
+    let comment_parser = comment_parser
+  end
+  in
   (module Matcher.Make (User_language) : Types.Matcher.S)

src/main.ml

@@ -138,23 +138,22 @@ let select_matcher custom_matcher override_matcher configuration =
       Yojson.Safe.from_file matcher_path
       |> Matchers.Syntax_config.of_yojson
       |> function
-      | Ok c -> Matchers.create c
+      | Ok c -> Matchers.create c, None
       | Error error ->
         Format.eprintf "%s@." error;
         exit 1
+  else if Option.is_some override_matcher then
+    Matchers.select_with_extension (Option.value_exn override_matcher), None
   else
-  if Option.is_some override_matcher then
-    Matchers.select_with_extension (Option.value_exn override_matcher)
-  else
-    match configuration.file_filters with
-    | None | Some [] ->
-      Matchers.select_with_extension ".generic"
-    | Some (filter::_) ->
-      match Filename.split_extension filter with
-      | _, Some extension ->
-        Matchers.select_with_extension ("." ^ extension)
-      | _ ->
-        Matchers.select_with_extension ".generic"
+    let extension =
+      match configuration.file_filters with
+      | None | Some [] -> ".generic"
+      | Some (filter::_) ->
+        match Filename.split_extension filter with
+        | _, Some extension -> "." ^ extension
+        | extension, None -> "." ^ extension
+    in
+    Matchers.select_with_extension extension, Some extension
 
 let write_statistics number_of_matches paths total_time dump_statistics =
   if dump_statistics then
@@ -414,10 +413,14 @@ let base_command_parameters : (unit -> 'result) Command.Param.t =
         Format.eprintf "%s@." @@ Error.to_string_hum error;
         exit 1
     in
-    let matcher = select_matcher custom_matcher override_matcher configuration
-    in
+    let matcher, extension = select_matcher custom_matcher override_matcher configuration in
     fun () ->
-      run matcher configuration
+      run matcher configuration;
+      match extension with
+      | Some ".generic" ->
+        Format.eprintf "WARNING: the GENERIC matcher was used, because a language could not be inferred from the file extension. Use something like '-matcher .go' (including the dot, '.') to force using the matcher for Go, and similarly '-matcher .java' for Java, and so on for other languages.@."
+      | Some _
+      | None -> ()
   ]
 
 let default_command =

test/test_cli.ml

@@ -210,9 +210,10 @@ let%expect_test "with_rewrite_rule_stdin_default_no_extension" =
     Format.sprintf "-sequential '%s' '%s' -rule '%s' -stdin" match_template rewrite_template rule
   in
   let command = Format.sprintf "%s %s" binary_path command_args in
-  let result = read_source_from_stdin command source in
+  let result = read_expect_stdin_and_stdout command source in
   print_string result;
-  [%expect_exact {|------ /dev/null
+  [%expect_exact {|WARNING: the GENERIC matcher was used, because a language could not be inferred from the file extension. Use something like '-matcher .go' (including the dot, '.') to force using the matcher for Go, and similarly '-matcher .java' for Java, and so on for other languages.
+------ /dev/null
 ++++++ /dev/null
 @|-1,1 +1,1 ============================================================
 !|hello world
@@ -226,9 +227,10 @@ let%expect_test "generic_matcher" =
     Format.sprintf "-stdin -sequential '%s' '%s' -f .generic" match_template rewrite_template
   in
   let command = Format.sprintf "%s %s" binary_path command_args in
-  let result = read_source_from_stdin command source in
+  let result = read_expect_stdin_and_stdout command source in
   print_string result;
-  [%expect_exact {|------ /dev/null
+  [%expect_exact {|WARNING: the GENERIC matcher was used, because a language could not be inferred from the file extension. Use something like '-matcher .go' (including the dot, '.') to force using the matcher for Go, and similarly '-matcher .java' for Java, and so on for other languages.
+------ /dev/null
 ++++++ /dev/null
 @|-1,1 +1,1 ============================================================
 -|\footnote{\small \url{https://github.com}}
@@ -422,9 +424,7 @@ let%expect_test "nested_templates" =
   let template_dir = "example" ^/ "multiple-nested-templates" in
   let command_args = Format.sprintf "-stdin -sequential -f .c -templates %s -stdout" template_dir in
   let command = Format.sprintf "%s %s" binary_path command_args in
-  let result =
-    read_expect_stdin_and_stdout command source
-  in
+  let result = read_expect_stdin_and_stdout command source in
   print_string result;
   [%expect{|
     Warning: Could not read required match file in example/multiple-nested-templates/invalid-subdir
@@ -720,9 +720,10 @@ let%expect_test "infer_and_honor_extensions" =
   let src_dir = "example" ^/ "src" ^/ "honor-file-extensions" in
   let command_args = Format.sprintf "'foo()' 'bar()' .generic -sequential -d %s -diff" src_dir in
   let command = Format.sprintf "%s %s" binary_path command_args in
-  let result = read_source_from_stdin command source in
+  let result = read_expect_stdin_and_stdout command source in
   print_string result;
   [%expect{|
+    WARNING: the GENERIC matcher was used, because a language could not be inferred from the file extension. Use something like '-matcher .go' (including the dot, '.') to force using the matcher for Go, and similarly '-matcher .java' for Java, and so on for other languages.
     --- example/src/honor-file-extensions/honor.pb.generic
     +++ example/src/honor-file-extensions/honor.pb.generic
     @@ -1,3 +1,3 @@
@@ -738,6 +739,8 @@ let%expect_test "diff_only" =
   let result = read_source_from_stdin command source in
   print_string result;
   [%expect{|
+    WARNING: the GENERIC matcher was used, because a language could not be inferred from the file extension. Use something like '-matcher .go' (including the dot, '.') to force using the matcher for Go, and similarly '-matcher .java' for Java, and so on for other languages.
+
     {"uri":null,"diff":"--- /dev/null\n+++ /dev/null\n@@ -1,1 +1,1 @@\n-hello world\n+world world"} |}];
 
   let source = "hello world" in
@@ -770,9 +773,10 @@ let%expect_test "zip_exclude_dir_no_extension" =
   let exclude_dir = "sample-repo/vendor" in
   let command_args = Format.sprintf "'main' 'pain' -zip %s -sequential -diff -exclude-dir %s" zip exclude_dir in
   let command = Format.sprintf "%s %s" binary_path command_args in
-  let result = read_source_from_stdin command source in
+  let result = read_expect_stdin_and_stdout command source in
   print_string result;
   [%expect{|
+    WARNING: the GENERIC matcher was used, because a language could not be inferred from the file extension. Use something like '-matcher .go' (including the dot, '.') to force using the matcher for Go, and similarly '-matcher .java' for Java, and so on for other languages.
     --- sample-repo/src/main.go
     +++ sample-repo/src/main.go
     @@ -1,2 +1,2 @@