feat: SP-2487 Implement path obfuscation

Author: isasmendiagus

src/main/java/com/scanoss/Scanner.java

@@ -49,6 +49,7 @@
 import java.util.concurrent.Executors;
 import java.util.concurrent.Future;
 import java.util.function.Predicate;
+import java.util.stream.Collectors;
 
 import static com.scanoss.ScanossConstants.*;
 
@@ -391,8 +392,16 @@ public List<String> scanFileList(@NonNull String folder, @NonNull List<String> f
      * @return Processed results, either modified based on BOM or original results if no BOM exists
      */
     private List<String> postProcessResults(List<String> results) {
+        List<ScanFileResult> scanFileResults = JsonUtils.toScanFileResults(results);
+
+        if (obfuscate) {
+            scanFileResults = scanFileResults.stream().map(s -> {
+                String originalFilePath = winnowing.getRealFilePathFor(s.getFilePath()).toString();
+                return s.withFilePath(originalFilePath);
+            }).collect(Collectors.toList());
+        }
+
         if (settings.getBom() != null) {
-            List<ScanFileResult> scanFileResults = JsonUtils.toScanFileResults(results);
             List <ScanFileResult>  newScanFileResults = this.postProcessor.process(scanFileResults, this.settings.getBom());
             return JsonUtils.toRawJsonString(newScanFileResults);
         }

src/main/java/com/scanoss/Winnowing.java

@@ -28,6 +28,7 @@
 import lombok.*;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.commons.codec.digest.DigestUtils;
+import org.apache.commons.io.FilenameUtils;
 import org.apache.tika.Tika;
 import org.apache.tika.mime.MediaType;
 import org.apache.tika.mime.MediaTypeRegistry;
@@ -37,6 +38,7 @@
 import java.io.IOException;
 import java.nio.charset.Charset;
 import java.nio.file.Files;
+import java.nio.file.Path;
 import java.util.*;
 import java.util.zip.CRC32C;
 import java.util.zip.Checksum;
@@ -69,6 +71,12 @@ public class Winnowing {
     @Builder.Default
     private int snippetLimit = MAX_LONG_LINE_CHARS; // Enable limiting of size of a single line of snippet generation
 
+    private Map<String, Path> obfuscationMap;
+
+    public Path getRealFilePathFor(String obfuscatedPath) {
+        return obfuscationMap.get(obfuscatedPath);
+    }
+
     /**
      * Calculate the WFP (fingerprint) for the given file
      *
@@ -112,7 +120,14 @@ public String wfpForContents(@NonNull String filename, Boolean binFile, byte[] c
         char[] fileContents = (new String(contents, Charset.defaultCharset())).toCharArray();
         String fileMD5 = DigestUtils.md5Hex(contents);
         StringBuilder wfpBuilder = new StringBuilder();
-        // TODO add obfuscation of the filename here
+
+        if (obfuscate) {
+            String extension = FilenameUtils.getExtension(filename);
+            String obfuscatedPath = obfuscationMap.size() + "." + extension;
+            this.obfuscationMap.put(obfuscatedPath, Path.of(filename));
+            filename = obfuscatedPath;
+        }
+
         wfpBuilder.append(String.format("file=%s,%d,%s\n", fileMD5, contents.length, filename));
         if (binFile || this.skipSnippets || this.skipSnippets(filename, fileContents)) {
             return wfpBuilder.toString();

src/main/java/com/scanoss/dto/ScanFileResult.java

@@ -23,6 +23,7 @@
 package com.scanoss.dto;
 
 import lombok.Data;
+import lombok.With;
 
 import java.util.List;
 
@@ -31,6 +32,7 @@
  */
 @Data
 public class ScanFileResult {
+    @With
     private final String filePath;
     private final List<ScanFileDetails> fileDetails;
 }