Remove tree-sitter-cpp submodule

.github/dependabot.yml

@@ -5,6 +5,11 @@ updates:
   schedule:
     interval: weekly
   open-pull-requests-limit: 99
+- package-ecosystem: cargo
+  directory: "/tree-sitter-mozcpp"
+  schedule:
+    interval: weekly
+  open-pull-requests-limit: 99
 - package-ecosystem: cargo
   directory: "/enums"
   schedule:

.gitmodules

@@ -10,6 +10,3 @@
 [submodule "tree-sitter-typescript"]
 	path = tree-sitter-typescript
 	url = https://github.com/tree-sitter/tree-sitter-typescript/
-[submodule "tree-sitter-cpp"]
-	path = tree-sitter-cpp
-	url = https://github.com/tree-sitter/tree-sitter-cpp.git

build.rs

@@ -160,7 +160,6 @@ fn main() {
         "tree-sitter-ccomment".to_string(),
         "tree-sitter-mozcpp".to_string(),
         "tree-sitter-typescript".to_string(),
-        "tree-sitter-cpp".to_string(),
     ];
     let dirs = collect_tree_sitter_dirs(ignore);
     for dir in dirs {

generate-moz-grammars/generate-mozcpp.sh

@@ -4,10 +4,8 @@
 #
 # Usage: ./generate-moz-grammars/generate-mozcpp.sh
 
-# FIXME we need to remove this line once we are going to use
-# the tree-sitter-cpp bindings
-# Get the tree-sitter-cpp submodule version
-TS_CPP_VERSION=`git submodule status tree-sitter-cpp | awk '{ print $1 }'`
+# Set tree-sitter-cpp version
+TS_CPP_VERSION="a35a275df92e7583df38f2de2562361f2b69987e"
 
 # Enter the mozcpp directory
 pushd tree-sitter-mozcpp
@@ -30,6 +28,9 @@ npm install -y
 # Exit tree-sitter-cpp directory
 popd
 
+# Copy tree-sitter-cpp `scanner.cc` functions into the `src` directory
+cp --verbose tree-sitter-cpp/src/scanner.cc ./src/scanner.cc
+
 # Init npm
 npm init -y
 

tree-sitter-mozcpp/Cargo.toml

@@ -25,3 +25,6 @@ tree-sitter = "^0.17"
 
 [build-dependencies]
 cc = "^1.0"
+# This dependency is not used at all for this crate, but it is here so that
+# dependabot can send notifications when there are updates for this grammar
+tree-sitter-cpp = "0.19.0"

tree-sitter-mozcpp/src/scanner.cc

@@ -1 +1,119 @@
-#include "../../tree-sitter-cpp/src/scanner.cc"
+#include <tree_sitter/parser.h>
+#include <string>
+#include <cwctype>
+
+namespace {
+
+using std::wstring;
+using std::iswspace;
+
+enum TokenType {
+  RAW_STRING_LITERAL,
+};
+
+struct Scanner {
+  bool scan(TSLexer *lexer, const bool *valid_symbols) {
+    while (iswspace(lexer->lookahead)) {
+      lexer->advance(lexer, true);
+    }
+
+    lexer->result_symbol = RAW_STRING_LITERAL;
+
+    // Raw string literals can start with: R, LR, uR, UR, u8R
+    // Consume 'R'
+    if (lexer->lookahead == 'L' || lexer->lookahead == 'U') {
+      lexer->advance(lexer, false);
+      if (lexer->lookahead != 'R') {
+        return false;
+      }
+    } else if (lexer->lookahead == 'u') {
+      lexer->advance(lexer, false);
+      if (lexer->lookahead == '8') {
+        lexer->advance(lexer, false);
+        if (lexer->lookahead != 'R') {
+          return false;
+        }
+      } else if (lexer->lookahead != 'R') {
+        return false;
+      }
+    } else if (lexer->lookahead != 'R') {
+      return false;
+    }
+    lexer->advance(lexer, false);
+
+    // Consume '"'
+    if (lexer->lookahead != '"') return false;
+    lexer->advance(lexer, false);
+
+    // Consume '(', delimiter
+    wstring delimiter;
+    for (;;) {
+      if (lexer->lookahead == 0 || lexer->lookahead == '\\' || iswspace(lexer->lookahead)) {
+        return false;
+      }
+      if (lexer->lookahead == '(') {
+        lexer->advance(lexer, false);
+        break;
+      }
+      delimiter += lexer->lookahead;
+      lexer->advance(lexer, false);
+    }
+
+    // Consume content, delimiter, ')', '"'
+    int delimiter_index = -1;
+    for (;;) {
+      if (lexer->lookahead == 0) return false;
+
+      if (delimiter_index >= 0) {
+        if (static_cast<unsigned>(delimiter_index) == delimiter.size()) {
+          if (lexer->lookahead == '"') {
+            lexer->advance(lexer, false);
+            return true;
+          } else {
+            delimiter_index = -1;
+          }
+        } else {
+          if (lexer->lookahead == delimiter[delimiter_index]) {
+            delimiter_index++;
+          } else {
+            delimiter_index = -1;
+          }
+        }
+      }
+
+      if (delimiter_index == -1 && lexer->lookahead == ')') {
+        delimiter_index = 0;
+      }
+
+      lexer->advance(lexer, false);
+    }
+  }
+};
+
+}
+
+extern "C" {
+
+void *tree_sitter_cpp_external_scanner_create() {
+  return new Scanner();
+}
+
+bool tree_sitter_cpp_external_scanner_scan(void *payload, TSLexer *lexer,
+                                            const bool *valid_symbols) {
+  Scanner *scanner = static_cast<Scanner *>(payload);
+  return scanner->scan(lexer, valid_symbols);
+}
+
+unsigned tree_sitter_cpp_external_scanner_serialize(void *payload, char *buffer) {
+  return 0;
+}
+
+void tree_sitter_cpp_external_scanner_deserialize(void *payload, const char *buffer, unsigned length) {
+}
+
+void tree_sitter_cpp_external_scanner_destroy(void *payload) {
+  Scanner *scanner = static_cast<Scanner *>(payload);
+  delete scanner;
+}
+
+}