Remove exif in image file (#2168)

* Issue-1790[feature] add pre handler for attachment files

* Issue-2121 remove exif from JPEG

* Issue-2121 add an 'enable' option in preHandler

* Issue-2121 add an 'enable' option in preHandler

* Issue-2121 remove useless code

* Issue-2121 ImageFileMultipartFile

* Issue-2121 change method name

* Issue-1720 remove useless annotation and add author on file

* Issue-1720
1. Modify for code review.
2. Keep 'Orientation' Exif tag for images.

* Issue-1720 for review

* Issue-1720 for review

* Issue-1720 remove all Exif

build.gradle

@@ -75,6 +75,7 @@ ext {
     jsoupVersion = '1.14.3'
     embeddedRedisVersion = '0.6'
     diffUtilsVersion = '4.11'
+    commonsImagingVersion = "1.0-alpha3"
 }
 
 dependencies {
@@ -125,6 +126,7 @@ dependencies {
     implementation "org.flywaydb:flyway-core:$flywayVersion"
     implementation "com.google.zxing:core:$zxingVersion"
     implementation "io.github.java-diff-utils:java-diff-utils:$diffUtilsVersion"
+    implementation "org.apache.commons:commons-imaging:$commonsImagingVersion"
 
     implementation "org.iq80.leveldb:leveldb:$levelDbVersion"
     runtimeOnly "com.h2database:h2:$h2Version"

src/main/java/run/halo/app/handler/file/FileHandlers.java

@@ -1,8 +1,10 @@
 package run.halo.app.handler.file;
 
+import java.io.IOException;
 import java.util.Collection;
 import java.util.concurrent.ConcurrentHashMap;
 import lombok.extern.slf4j.Slf4j;
+import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.context.ApplicationContext;
 import org.springframework.lang.NonNull;
 import org.springframework.lang.Nullable;
@@ -10,8 +12,11 @@ import org.springframework.stereotype.Component;
 import org.springframework.util.Assert;
 import org.springframework.util.CollectionUtils;
 import org.springframework.web.multipart.MultipartFile;
+import run.halo.app.exception.BadRequestException;
 import run.halo.app.exception.FileOperationException;
 import run.halo.app.exception.RepeatTypeException;
+import run.halo.app.handler.prehandler.ByteMultipartFile;
+import run.halo.app.handler.prehandler.FilePreHandlers;
 import run.halo.app.model.entity.Attachment;
 import run.halo.app.model.enums.AttachmentType;
 import run.halo.app.model.support.UploadResult;
@@ -26,6 +31,9 @@ import run.halo.app.model.support.UploadResult;
 @Component
 public class FileHandlers {
 
+    @Autowired
+    private FilePreHandlers filePreHandlers;
+
     /**
      * File handler container.
      */
@@ -49,7 +57,14 @@ public class FileHandlers {
      */
     @NonNull
     public UploadResult upload(@NonNull MultipartFile file,
-        @NonNull AttachmentType attachmentType) {
+                               @NonNull AttachmentType attachmentType) {
+        try {
+            byte[] bytes = filePreHandlers.process(file.getBytes());
+            file = new ByteMultipartFile(bytes, file.getOriginalFilename(), file.getName(),
+                file.getContentType());
+        } catch (IOException e) {
+            throw new BadRequestException("Get file bytes for preprocess failed", e);
+        }
         return getSupportedType(attachmentType).upload(file);
     }
 
@@ -94,4 +109,5 @@ public class FileHandlers {
         }
         return handler;
     }
+
 }

src/main/java/run/halo/app/handler/prehandler/ByteMultipartFile.java

@@ -0,0 +1,69 @@
+package run.halo.app.handler.prehandler;
+
+import java.io.ByteArrayInputStream;
+import java.io.File;
+import java.io.IOException;
+import java.io.InputStream;
+import org.jetbrains.annotations.NotNull;
+import org.springframework.util.FileCopyUtils;
+import org.springframework.web.multipart.MultipartFile;
+
+/**
+ * @author eziosudo
+ * @date 2022-06-19
+ */
+public class ByteMultipartFile implements MultipartFile {
+
+    private final byte[] imgContent;
+    private final String originalFilename;
+    private final String name;
+    private final String contentType;
+
+    public ByteMultipartFile(byte[] imgContent, String originalFilename, String name,
+                             String contentType) {
+        this.imgContent = imgContent;
+        this.originalFilename = originalFilename;
+        this.name = name;
+        this.contentType = contentType;
+    }
+
+    @Override
+    public String getName() {
+        return this.name;
+    }
+
+    @Override
+    public String getOriginalFilename() {
+        return this.originalFilename;
+    }
+
+    @Override
+    public String getContentType() {
+        return this.contentType;
+    }
+
+    @Override
+    public boolean isEmpty() {
+        return imgContent == null || imgContent.length == 0;
+    }
+
+    @Override
+    public long getSize() {
+        return imgContent.length;
+    }
+
+    @Override
+    public byte[] getBytes() throws IOException {
+        return imgContent;
+    }
+
+    @Override
+    public InputStream getInputStream() throws IOException {
+        return new ByteArrayInputStream(imgContent);
+    }
+
+    @Override
+    public void transferTo(@NotNull File dest) throws IOException, IllegalStateException {
+        FileCopyUtils.copy(imgContent, dest);
+    }
+}

src/main/java/run/halo/app/handler/prehandler/FilePreHandler.java

@@ -0,0 +1,11 @@
+package run.halo.app.handler.prehandler;
+
+/**
+ * @author eziosudo
+ * @date 2022-06-16
+ */
+public interface FilePreHandler {
+
+    byte[] preProcess(byte[] bytes);
+
+}

src/main/java/run/halo/app/handler/prehandler/FilePreHandlers.java

@@ -0,0 +1,37 @@
+package run.halo.app.handler.prehandler;
+
+import java.util.List;
+import java.util.stream.Collectors;
+import org.springframework.context.ApplicationContext;
+import org.springframework.stereotype.Component;
+
+/**
+ * @author eziosudo
+ * @date 2022-06-16
+ */
+@Component
+public class FilePreHandlers {
+
+    private final List<FilePreHandler> preHandlers;
+
+    FilePreHandlers(ApplicationContext applicationContext) {
+        preHandlers =
+            applicationContext.getBeanProvider(FilePreHandler.class)
+                .orderedStream()
+                .collect(Collectors.toList());
+    }
+
+    /**
+     * 遍历预处理方法，对输入文件进行预处理
+     *
+     * @param bytes 输入文件字节流
+     * @return 输出预处理后的字节流
+     */
+    public byte[] process(byte[] bytes) {
+        for (FilePreHandler filePreHandler : preHandlers) {
+            bytes = filePreHandler.preProcess(bytes);
+        }
+        return bytes;
+    }
+
+}

src/main/java/run/halo/app/handler/prehandler/PictureExifRemovalPreHandler.java

@@ -0,0 +1,85 @@
+package run.halo.app.handler.prehandler;
+
+import java.io.BufferedOutputStream;
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;
+import java.io.OutputStream;
+import lombok.extern.slf4j.Slf4j;
+import org.apache.commons.imaging.ImageReadException;
+import org.apache.commons.imaging.ImageWriteException;
+import org.apache.commons.imaging.Imaging;
+import org.apache.commons.imaging.common.ImageMetadata;
+import org.apache.commons.imaging.formats.jpeg.JpegImageMetadata;
+import org.apache.commons.imaging.formats.jpeg.exif.ExifRewriter;
+import org.apache.commons.imaging.formats.tiff.TiffImageMetadata;
+import org.apache.commons.imaging.formats.tiff.write.TiffOutputDirectory;
+import org.apache.commons.imaging.formats.tiff.write.TiffOutputField;
+import org.apache.commons.imaging.formats.tiff.write.TiffOutputSet;
+import org.apache.commons.lang3.StringUtils;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.core.annotation.Order;
+import org.springframework.stereotype.Component;
+import run.halo.app.model.properties.AttachmentProperties;
+import run.halo.app.service.OptionService;
+
+/**
+ * @author eziosudo
+ * @date 2022-06-16
+ */
+@Slf4j
+@Component
+@Order(value = 1)
+public class PictureExifRemovalPreHandler implements FilePreHandler {
+
+    @Autowired
+    private OptionService optionService;
+
+    @Override
+    public byte[] preProcess(byte[] bytes) {
+        if (!isRemoveExifEnable()) {
+            return bytes;
+        }
+        try {
+            ImageMetadata metadata = Imaging.getMetadata(bytes);
+            final JpegImageMetadata jpegMetadata = (JpegImageMetadata) metadata;
+            if (null == jpegMetadata) {
+                return bytes;
+            }
+            final TiffImageMetadata exif = jpegMetadata.getExif();
+            if (null == exif) {
+                return bytes;
+            }
+            final TiffOutputSet outputSet = exif.getOutputSet();
+            if (null == outputSet) {
+                return bytes;
+            }
+            try (ByteArrayOutputStream bos = new ByteArrayOutputStream();
+                 OutputStream os = new BufferedOutputStream(bos)) {
+                for (TiffOutputDirectory directory : outputSet.getDirectories()) {
+                    for (TiffOutputField field : directory.getFields()) {
+                        if (!StringUtils.equalsAnyIgnoreCase("Orientation", field.tagInfo.name)) {
+                            outputSet.removeField(field.tagInfo);
+                        }
+                    }
+                }
+                new ExifRewriter().updateExifMetadataLossless(bytes, os, outputSet);
+                bytes = bos.toByteArray();
+                return bytes;
+            }
+        } catch (IllegalArgumentException e) {
+            log.info("Cannot parse to image format.");
+        } catch (ImageWriteException | ImageReadException | IOException e) {
+            log.info("Cannot get metadata from bytes.", e);
+        } catch (Exception e) {
+            log.info("Cannot check or remove Exif from bytes.", e);
+        }
+        return bytes;
+    }
+
+    private boolean isRemoveExifEnable() {
+        return optionService.getByPropertyOrDefault(
+            AttachmentProperties.REMOVE_IMAGE_EXIF_ENABLE,
+            Boolean.class, false);
+    }
+
+}

src/main/java/run/halo/app/model/properties/AttachmentProperties.java

@@ -15,6 +15,11 @@ public enum AttachmentProperties implements PropertyEnum {
      */
     UPLOAD_IMAGE_PREVIEW_ENABLE("attachment_upload_image_preview_enable", Boolean.class, "true"),
 
+    /**
+     * Remove Exif in image when upload
+     */
+    REMOVE_IMAGE_EXIF_ENABLE("attachment_EXIF_remove_enable", Boolean.class, "false"),
+
     /**
      * Upload max parallel uploads
      */