feat: add tree-sitter support for Java

Signed-off-by: Joseph Kato <joseph@jdkato.io>

Author: jdkato

internal/core/format.go

@@ -68,7 +68,7 @@ var FormatByExtension = map[string][]string{
 	`\.(?:go)$`:                       {".go", "code"},
 	`\.(?:hs)$`:                       {".hs", "code"},
 	`\.(?:html|htm|shtml|xhtml)$`:     {".html", "markup"},
-	`\.(?:java|bsh)$`:                 {".c", "code"},
+	`\.(?:java|bsh)$`:                 {".java", "code"},
 	`\.(?:jl)$`:                       {".jl", "code"},
 	`\.(?:js|jsx)$`:                   {".js", "code"},
 	`\.(?:lua)$`:                      {".lua", "code"},

internal/lint/code/java.go

@@ -0,0 +1,20 @@
+package code
+
+import (
+	"regexp"
+
+	"github.com/errata-ai/vale/v3/internal/core"
+	"github.com/smacker/go-tree-sitter/java"
+)
+
+func Java() *Language {
+	return &Language{
+		Delims: regexp.MustCompile(`//|/\*|\*/`),
+		Parser: java.GetLanguage(),
+		Queries: []core.Scope{
+			{Name: "", Expr: "(line_comment)+ @comment", Type: ""},
+			{Name: "", Expr: "(block_comment)+ @comment", Type: ""},
+		},
+		Padding: cStyle,
+	}
+}

internal/lint/code/lang.go

@@ -40,6 +40,8 @@ func GetLanguageFromExt(ext string) (*Language, error) {
 		return JavaScript(), nil
 	case ".jl":
 		return Julia(), nil
+	case ".java":
+		return Java(), nil
 	case ".ts":
 		return TypeScript(), nil
 	case ".tsx":

testdata/features/lint.feature

@@ -34,6 +34,14 @@ Feature: Lint
             test.jl:47:3:vale.Annotations:'XXX' left in text
             """
 
+    Scenario: Lint a Java file
+        When I lint "test.java"
+        Then the output should contain exactly:
+            """
+            test.java:1:4:vale.Annotations:'XXX' left in text
+            test.java:9:4:vale.Annotations:'TODO' left in text
+            """
+
     Scenario: Lint a proto file
         When I lint "test.proto"
         Then the output should contain exactly:

testdata/fixtures/formats/test.java

@@ -0,0 +1,15 @@
+// XXX: This is an annotation
+// And this uses too many weasel words like obviously and very.
+
+/*
+ * This is a multi-line comment.
+ * It can span across several lines.
+ * All text within these delimiters is ignored.
+ *
+ * TODO: Refactor this code to improve readability.x
+ */
+public class HelloWorld {
+  public static void main(String[] args) {
+    System.out.println("Hello World, XXX!");
+  }
+}