Use only cloc.pl on both Unix and Windows

.github/workflows/ci.yml

@@ -39,6 +39,8 @@ jobs:
     steps:
       - name: Checkout latest
         uses: actions/checkout@v4
+      - name: Set up Perl
+        uses: shogo82148/actions-setup-perl@v1.31.4
       - name: Set up JDK 8
         uses: actions/setup-java@v4
         with:
@@ -55,6 +57,8 @@ jobs:
     steps:
       - name: Checkout latest
         uses: actions/checkout@v4
+      - name: Set up Perl
+        uses: shogo82148/actions-setup-perl@v1.31.4
       - name: Set up JDK 8
         uses: actions/setup-java@v4
         with:

README.md

@@ -14,8 +14,12 @@ following dependency in your Maven project:
 ## Requirements
 
 This library requires a minimum of Java 8. You don't need to have `cloc` installed on your system, as the JAR comes
-bundled with the necessary script. Since it is written in Perl, it should be executable out of the box on most systems
-(as the majority of Unix-like systems have it installed by default).
+bundled with the necessary script.
+
+> [!NOTE]  
+> Since this script is written in Perl, you must make sure that its interpreter is installed and located on your `PATH`.
+> Should work out of the box on most systems (as the majority of Unix-like systems have it installed by default),
+> but worth pointing out in case you plan on using this on a minimalistic Linux image or Windows.
 
 ## Example
 

doc/README.md

@@ -14,8 +14,12 @@ following dependency in your Maven project:
 ## Requirements
 
 This library requires a minimum of Java 8. You don't need to have `cloc` installed on your system, as the JAR comes
-bundled with the necessary script. Since it is written in Perl, it should be executable out of the box on most systems
-(as the majority of Unix-like systems have it installed by default).
+bundled with the necessary script.
+
+> [!NOTE]  
+> Since this script is written in Perl, you must make sure that its interpreter is installed and located on your `PATH`.
+> Should work out of the box on most systems (as the majority of Unix-like systems have it installed by default),
+> but worth pointing out in case you plan on using this on a minimalistic Linux image or Windows.
 
 ## Example
 

pom.xml

@@ -62,6 +62,7 @@
     <java.version>8</java.version>
     <cloc.version>2.02</cloc.version>
     <cloc.url>https://github.com/AlDanial/cloc</cloc.url>
+    <cloc.md5>31533dd445610364154ec328bfcf7b2c</cloc.md5>
     <org.apache.commons-lang.version>3.17.0</org.apache.commons-lang.version>
     <org.codehaus.plexus.version>4.0.2</org.codehaus.plexus.version>
     <org.jetbrains.annotations.version>26.0.1</org.jetbrains.annotations.version>
@@ -297,19 +298,7 @@
             <configuration>
               <url>${cloc.url}/releases/download/v${cloc.version}/cloc-${cloc.version}.pl</url>
               <outputFileName>cloc.pl</outputFileName>
-              <md5>31533dd445610364154ec328bfcf7b2c</md5>
-            </configuration>
-          </execution>
-          <execution>
-            <id>default-download-exe</id>
-            <phase>initialize</phase>
-            <goals>
-              <goal>wget</goal>
-            </goals>
-            <configuration>
-              <url>${cloc.url}/releases/download/v${cloc.version}/cloc-${cloc.version}.exe</url>
-              <outputFileName>cloc.exe</outputFileName>
-              <md5>31adac93c8d3c9e4d67912a8f9f5b729</md5>
+              <md5>${cloc.md5}</md5>
             </configuration>
           </execution>
         </executions>

src/main/java/ch/usi/si/seart/cloc/CLOC.java

@@ -19,6 +19,7 @@
 import java.io.UncheckedIOException;
 import java.net.URL;
 import java.nio.file.Path;
+import java.security.NoSuchAlgorithmException;
 import java.util.LinkedHashMap;
 import java.util.LinkedHashSet;
 import java.util.Map;
@@ -36,10 +37,6 @@
  */
 public final class CLOC {
 
-    private static final String CMD = "cloc";
-
-    private static final String EXECUTABLE = getBundledExecutable();
-
     private static final JsonMapper DEFAULT_MAPPER = new JsonMapper();
     private static volatile JsonMapper OUTPUT_MAPPER = DEFAULT_MAPPER;
 
@@ -67,6 +64,46 @@ private static Properties getProperties() {
         return PROPERTIES;
     }
 
+    private static volatile File EXECUTABLE;
+    private static File getExecutable() {
+        if (EXECUTABLE == null) {
+            synchronized (CLOC.class) {
+                if (EXECUTABLE == null) {
+                    URL url = CLOC.class.getClassLoader().getResource("cloc.pl");
+                    String protocol = Objects.requireNonNull(url).getProtocol();
+                    switch (protocol) {
+                        case "file":
+                            String path = url.getPath();
+                            EXECUTABLE = new File(path);
+                            break;
+                        case "jar":
+                            try {
+                                File script = new File(SystemUtils.USER_HOME, "cloc.pl");
+                                if (getMD5() != null && script.exists() && MD5.hash(script).equals(getMD5())) {
+                                    EXECUTABLE = script;
+                                    break;
+                                }
+                                FileUtils.copyURLToFile(url, script);
+                                boolean success = script.setExecutable(true);
+                                if (success) {
+                                    EXECUTABLE = script;
+                                    break;
+                                }
+                                throw new IOException("Unable change execute permissions: " + script);
+                            } catch (IOException ex) {
+                                throw new UncheckedIOException(ex);
+                            } catch (NoSuchAlgorithmException ex) {
+                                throw new IllegalStateException(ex);
+                            }
+                        default:
+                            throw new UnsupportedOperationException("Unsupported protocol: " + protocol);
+                    }
+                }
+            }
+        }
+        return EXECUTABLE;
+    }
+
     /**
      * @return the {@code cloc} repository link, or {@code null} if the details could not be loaded.
      */
@@ -83,6 +120,14 @@ public static String getVersion() {
         return getProperties().getProperty("cloc.version");
     }
 
+    /**
+     * @return the {@code cloc} executable MD5 checksum, or {@code null} if the details could not be loaded.
+     */
+    @Nullable
+    public static String getMD5() {
+        return getProperties().getProperty("cloc.md5");
+    }
+
     /**
      * Set the {@link JsonMapper} to use for parsing the output of the command.
      *
@@ -101,32 +146,6 @@ public static Builder command() {
         return new Builder();
     }
 
-    private static String getBundledExecutable() {
-        String extension = SystemUtils.IS_OS_WINDOWS ? "exe" : "pl";
-        URL url = CLOC.class.getClassLoader().getResource(CMD + "." + extension);
-        String protocol = Objects.requireNonNull(url).getProtocol();
-        switch (protocol) {
-            case "file":
-                return new File(url.getFile()).getPath();
-            case "jar":
-                try {
-                    File script = new File(SystemUtils.JAVA_IO_TMPDIR, CMD);
-                    FileUtils.copyURLToFile(url, script);
-                    boolean success = script.setExecutable(true);
-                    if (success) {
-                        script.deleteOnExit();
-                        return script.getAbsolutePath();
-                    } else {
-                        throw new IOException("Unable change execute permissions: " + script.getAbsolutePath());
-                    }
-                } catch (IOException ex) {
-                    throw new UncheckedIOException(ex);
-                }
-            default:
-                throw new UnsupportedOperationException("Unsupported protocol: " + protocol);
-        }
-    }
-
     /**
      * Facilitates the construction of {@link CLOC} command instances.
      * It allows for the step-by-step creation of these objects
@@ -270,12 +289,13 @@ public Builder skipUniqueness(boolean value) {
          */
         @Contract("_ -> new")
         public @NotNull CLOC target(@NotNull Path path) {
-            Objects.requireNonNull(path, "Path must not be null!");
-            File file = path.toFile();
+            File executable = getExecutable();
+            File file = Objects.requireNonNull(path, "Path must not be null!").toFile();
             if (!file.exists()) throw new IllegalArgumentException("Unable to read: " + path);
 
             CommandLine commandLine = new CommandLine();
-            commandLine.setExecutable(EXECUTABLE);
+            commandLine.createArg("perl");
+            commandLine.createArg().setFile(executable);
             commandLine.createArg().setFile(file);
             flags.stream()
                     .map(flag -> "--" + flag)

src/main/java/ch/usi/si/seart/cloc/MD5.java

@@ -0,0 +1,29 @@
+package ch.usi.si.seart.cloc;
+
+import java.io.File;
+import java.io.IOException;
+import java.io.InputStream;
+import java.math.BigInteger;
+import java.nio.file.Files;
+import java.security.DigestInputStream;
+import java.security.MessageDigest;
+import java.security.NoSuchAlgorithmException;
+
+class MD5 {
+
+    private MD5() {
+    }
+
+    @SuppressWarnings({"checkstyle:EmptyStatement", "StatementWithEmptyBody"})
+    static String hash(File file) throws IOException, NoSuchAlgorithmException {
+        try (
+            InputStream fileStream = Files.newInputStream(file.toPath());
+            DigestInputStream digestStream = new DigestInputStream(fileStream, MessageDigest.getInstance("MD5"))
+        ) {
+            byte[] buffer = new byte[8192];
+            while (digestStream.read(buffer, 0, buffer.length) != -1);
+            byte[] digested = digestStream.getMessageDigest().digest();
+            return new BigInteger(1, digested).toString(16);
+        }
+    }
+}

src/main/resources/cloc.properties

@@ -1,2 +1,3 @@
 cloc.url=${cloc.url}
 cloc.version=${cloc.version}
+cloc.md5=${cloc.md5}

src/test/java/ch/usi/si/seart/cloc/CLOCTest.java

@@ -93,4 +93,11 @@ void testGetVersion() {
         Assertions.assertNotNull(version, "Version should not be null.");
         Assertions.assertFalse(version.isEmpty(), "Version should not be empty.");
     }
+
+    @Test
+    void testGetMD5() {
+        String md5 = CLOC.getMD5();
+        Assertions.assertNotNull(md5, "MD5 should not be null.");
+        Assertions.assertFalse(md5.isEmpty(), "MD5 should not be empty.");
+    }
 }