Switched to SHA1 for the file hashing. This costs ms of time but ensures we won't get colliksions. Also is better distributed than md5

engine/src/main/resources/hibernate/ImageHash.hbm.xml

@@ -13,6 +13,6 @@
     <property name="ahash" column="ahash" type="long"/>
     <property name="dhash" column="dhash" type="long"/>
     <property name="phash" column="phash" type="long"/>
-    <property name="md5" column="md5" type="string"/>
+    <property name="fileHash" column="fileHash" type="string"/>
   </class>
 </hibernate-mapping>

engine/src/main/scala/com/sothr/imagetools/engine/dto/ImageHashDTO.scala

@@ -6,7 +6,7 @@ import grizzled.slf4j.Logging
 
 @Entity
 @Table(name = "ImageHash")
-class ImageHashDTO(var ahash: Long, var dhash: Long, var phash: Long, var md5: String) extends Serializable with Logging {
+class ImageHashDTO(var ahash: Long, var dhash: Long, var phash: Long, var fileHash: String) extends Serializable with Logging {
 
   @Id
   @GeneratedValue(strategy = GenerationType.AUTO)
@@ -38,14 +38,14 @@ class ImageHashDTO(var ahash: Long, var dhash: Long, var phash: Long, var md5: S
     phash = hash
   }
 
-  def getMd5: String = md5
+  def getFileHash: String = fileHash
 
-  def setMd5(hash: String) = {
-    md5 = hash
+  def setFileHash(hash: String) = {
+    fileHash = hash
   }
 
   def cloneHashes: ImageHashDTO = {
-    new ImageHashDTO(ahash, dhash, phash, md5)
+    new ImageHashDTO(ahash, dhash, phash, fileHash)
   }
 
   override def hashCode(): Int = {
@@ -57,6 +57,6 @@ class ImageHashDTO(var ahash: Long, var dhash: Long, var phash: Long, var md5: S
   }
 
   override def toString: String = {
-    s"MD5: $md5 ahash: $ahash dhash: $dhash phash: $phash"
+    s"fileHash: $fileHash ahash: $ahash dhash: $dhash phash: $phash"
   }
 }

engine/src/main/scala/com/sothr/imagetools/engine/hash/HashService.scala

@@ -29,7 +29,7 @@ object HashService extends Logging {
     var ahash: Long = 0L
     var dhash: Long = 0L
     var phash: Long = 0L
-    val md5: String = getMD5(imagePath)
+    val sha1: String = getSHA1(imagePath)
 
     //Get Image Data
     val grayImage = ImageService.convertToGray(image)
@@ -44,7 +44,7 @@ object HashService extends Logging {
       phash = getPhash(grayImage, alreadyGray = true)
     }
 
-    val hashes = new ImageHashDTO(ahash, dhash, phash, md5)
+    val hashes = new ImageHashDTO(ahash, dhash, phash, sha1)
     debug(s"Generated hashes: $hashes")
 
     hashes
@@ -91,8 +91,13 @@ object HashService extends Logging {
 
   def getMD5(filePath: String): String = {
     managed(new FileInputStream(filePath)) acquireAndGet {
-      input =>
-        DigestUtils.md5Hex(input)
+      input => DigestUtils.md5Hex(input)
+    }
+  }
+
+  def getSHA1(filePath: String): String = {
+    managed(new FileInputStream(filePath)) acquireAndGet {
+      input => DigestUtils.sha1Hex(input)
     }
   }
 

engine/src/main/scala/com/sothr/imagetools/engine/image/ImageService.scala

@@ -28,8 +28,8 @@ object ImageService extends Logging {
         debug(s"Processing image: ${file.getAbsolutePath}")
         val bufferedImage = ImageIO.read(file)
         val hashes = HashService.getImageHashes(bufferedImage, file.getAbsolutePath)
-        var thumbnailPath = lookupThumbnailPath(hashes.md5)
-        if (thumbnailPath == null) thumbnailPath = getThumbnail(bufferedImage, hashes.md5)
+        var thumbnailPath = lookupThumbnailPath(hashes.getFileHash)
+        if (thumbnailPath == null) thumbnailPath = getThumbnail(bufferedImage, hashes.getFileHash)
         val imageSize = {
           (bufferedImage.getWidth, bufferedImage.getHeight)
         }

engine/src/main/scala/com/sothr/imagetools/engine/util/Timing.scala

@@ -8,7 +8,7 @@ trait Timing extends Logging {
     val t0 = System.currentTimeMillis
     val result = block // call-by-name
     val t1 = System.currentTimeMillis
-    info("Elapsed time: " + (t1 - t0) + "ms")
+    debug("Elapsed time: " + (t1 - t0) + "ms")
     result
   }
 
@@ -16,7 +16,7 @@ trait Timing extends Logging {
     val t0 = System.currentTimeMillis
     val result = block // call-by-name
     val t1 = System.currentTimeMillis
-    info("Elapsed time: " + (t1 - t0) + "ms")
+    debug("Elapsed time: " + (t1 - t0) + "ms")
     t1 - t0
   }
 

engine/src/test/resources/hibernate/ImageHash.hbm.xml

@@ -13,6 +13,6 @@
     <property name="ahash" column="ahash" type="long"/>
     <property name="dhash" column="dhash" type="long"/>
     <property name="phash" column="phash" type="long"/>
-    <property name="md5" column="md5" type="string"/>
+    <property name="fileHash" column="fileHash" type="string"/>
   </class>
 </hibernate-mapping>

engine/src/test/resources/logback-minimum-config.xml

@@ -51,7 +51,7 @@
       <MaxFileSize>500KB</MaxFileSize>
     </triggeringPolicy>
   </appender>
-  <root level="DEBUG">
+  <root level="INFO">
     <appender-ref ref="DL"/>
     <appender-ref ref="IL"/>
     <appender-ref ref="EL"/>

engine/src/test/scala/com/sothr/imagetools/engine/hash/HashServiceTest.scala

@@ -335,6 +335,64 @@ class HashServiceTest extends BaseTest {
     assert(hash == "b137131bd55896c747286e4d247b845e")
   }
 
+  def sha1TestCase(filePath: String): String = {
+    HashService.getSHA1(filePath)
+  }
+
+  test("Benchmark SHA1") {
+    info("Benchmarking SHA1")
+    info("SHA1 Large Image 3684x2736")
+    val time = new mutable.MutableList[Long]()
+    for (runNum <- 0 until benchmarkRuns) {
+      time += getTime {
+        sha1TestCase(TestParams.LargeSampleImage1)
+      }
+    }
+    val largeMean = getMean(time.toArray[Long])
+    info(s"The mean time of ${time.size} tests for large was: $largeMean ms")
+    time.clear()
+    info("SHA1 Medium Image 1824x1368")
+    for (runNum <- 0 until benchmarkRuns) {
+      time += getTime {
+        sha1TestCase(TestParams.MediumSampleImage1)
+      }
+    }
+    val mediumMean = getMean(time.toArray[Long])
+    info(s"The mean time of ${time.size} tests for medium was: $mediumMean ms")
+    time.clear()
+    info("SHA1 Small Image 912x684")
+    for (runNum <- 0 until benchmarkRuns) {
+      time += getTime {
+        sha1TestCase(TestParams.SmallSampleImage1)
+      }
+    }
+    val smallMean = getMean(time.toArray[Long])
+    info(s"The mean time of ${time.size} tests for small was: $smallMean ms")
+    time.clear()
+    assert(true)
+  }
+
+  test("Calculate SHA1 Large Sample Image 1") {
+    debug("Starting 'Calculate SHA1 Large Sample Image 1' test")
+    val hash = HashService.getSHA1(TestParams.LargeSampleImage1)
+    debug(s"Testing that $hash = 4beb6f2d852b75a313863916a1803ebad13a3196")
+    assert(hash == "4beb6f2d852b75a313863916a1803ebad13a3196")
+  }
+
+  test("Calculate SHA1 Medium Sample Image 1") {
+    debug("Starting 'Calculate SHA1 Medium Sample Image 1' test")
+    val hash = HashService.getSHA1(TestParams.MediumSampleImage1)
+    debug(s"Testing that $hash = edc718ce8e3556a39592ffdc214d0f636529be9f")
+    assert(hash == "edc718ce8e3556a39592ffdc214d0f636529be9f")
+  }
+
+  test("Calculate SHA1 Small Sample Image 1") {
+    debug("Starting 'Calculate SHA1 Small Sample Image 1' test")
+    val hash = HashService.getSHA1(TestParams.SmallSampleImage1)
+    debug(s"Testing that $hash = 1a91d2b5327f0aad258419f76b87d4c0bc343443")
+    assert(hash == "1a91d2b5327f0aad258419f76b87d4c0bc343443")
+  }
+
   def imageHashTestWithCacheCase(filePath: String): ImageHashDTO = {
     val cache = AppConfig.cacheManager.getCache("images")
     var result: ImageHashDTO = null