feat: rewrite the scanner in C

.clang-format

@@ -0,0 +1,12 @@
+AlignArrayOfStructures: Left
+BasedOnStyle: LLVM
+IndentCaseLabels: true
+IndentGotoLabels: true
+IndentPPDirectives: AfterHash
+IndentWidth: 2
+KeepEmptyLinesAtTheStartOfBlocks: false
+SeparateDefinitionBlocks: Always
+SortIncludes: CaseInsensitive
+SpaceAfterCStyleCast: false
+SpaceAfterLogicalNot: false
+SpaceBeforeCaseColon: false

.gitattributes

@@ -1,4 +1,4 @@
 /src/**/* linguist-generated=true
-/src/scanner.cc linguist-generated=false
+/src/scanner.c linguist-generated=false
 # Exclude test files from language stats
 /test/**/*.ex linguist-documentation=true

Package.swift

@@ -30,7 +30,7 @@ let package = Package(
                 ],
                 sources: [
                     "src/parser.c",
-                    "src/scanner.cc",
+                    "src/scanner.c",
                 ],
                 resources: [
                     .copy("queries")

binding.gyp

@@ -9,7 +9,7 @@
       "sources": [
         "bindings/node/binding.cc",
         "src/parser.c",
-        "src/scanner.cc"
+        "src/scanner.c"
       ],
       "cflags_c": [
         "-std=c99",

bindings/rust/build.rs

@@ -2,34 +2,15 @@ fn main() {
     let src_dir = std::path::Path::new("src");
 
     let mut c_config = cc::Build::new();
-    c_config.include(&src_dir);
+    c_config.include(src_dir);
     c_config
         .flag_if_supported("-Wno-unused-parameter")
         .flag_if_supported("-Wno-unused-but-set-variable")
         .flag_if_supported("-Wno-trigraphs");
     let parser_path = src_dir.join("parser.c");
     c_config.file(&parser_path);
 
-    // If your language uses an external scanner written in C,
-    // then include this block of code:
-
-    /*
     let scanner_path = src_dir.join("scanner.c");
     c_config.file(&scanner_path);
     println!("cargo:rerun-if-changed={}", scanner_path.to_str().unwrap());
-    */
-
-    c_config.compile("parser");
-    println!("cargo:rerun-if-changed={}", parser_path.to_str().unwrap());
-
-    let mut cpp_config = cc::Build::new();
-    cpp_config.cpp(true);
-    cpp_config.include(&src_dir);
-    cpp_config
-        .flag_if_supported("-Wno-unused-parameter")
-        .flag_if_supported("-Wno-unused-but-set-variable");
-    let scanner_path = src_dir.join("scanner.cc");
-    cpp_config.file(&scanner_path);
-    cpp_config.compile("scanner");
-    println!("cargo:rerun-if-changed={}", scanner_path.to_str().unwrap());
 }

docs/parser.md

@@ -33,7 +33,7 @@ For detailed introduction see the official guide on [Creating parsers](https://t
 
 Essentially, we define relevant language rules in `grammar.js`, based on which
 Tree-sitter generates parser code (under `src/`). In some cases, we want to write
-custom C++ code for tokenizing specific character sequences (in `src/scanner.cc`).
+custom C code for tokenizing specific character sequences (in `src/scanner.c`).
 
 The grammar rules may often conflict with each other, meaning that the given
 sequence of tokens has multiple valid interpretations given one _token_ of lookahead.