File ImageFilter.java
Coverage histogram
0% of files have more coverage
Classes
| Class | Line # | Actions | |||||
|---|---|---|---|---|---|---|---|
| ImageFilter | 74 | 45 | 0% | 16 | 4 |
Contributing tests
This file is covered by 24 tests.
.
Source view
| 1 | /* | |
| 2 | * See the NOTICE file distributed with this work for additional | |
| 3 | * information regarding copyright ownership. | |
| 4 | * | |
| 5 | * This is free software; you can redistribute it and/or modify it | |
| 6 | * under the terms of the GNU Lesser General Public License as | |
| 7 | * published by the Free Software Foundation; either version 2.1 of | |
| 8 | * the License, or (at your option) any later version. | |
| 9 | * | |
| 10 | * This software is distributed in the hope that it will be useful, | |
| 11 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
| 12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
| 13 | * Lesser General Public License for more details. | |
| 14 | * | |
| 15 | * You should have received a copy of the GNU Lesser General Public | |
| 16 | * License along with this software; if not, write to the Free | |
| 17 | * Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA | |
| 18 | * 02110-1301 USA, or see the FSF site: http://www.fsf.org. | |
| 19 | */ | |
| 20 | package org.xwiki.officeimporter.internal.filter; | |
| 21 | ||
| 22 | import java.net.URLDecoder; | |
| 23 | import java.nio.charset.Charset; | |
| 24 | import java.util.HashMap; | |
| 25 | import java.util.List; | |
| 26 | import java.util.Map; | |
| 27 | ||
| 28 | import javax.inject.Inject; | |
| 29 | import javax.inject.Named; | |
| 30 | import javax.inject.Singleton; | |
| 31 | ||
| 32 | import org.apache.commons.lang3.StringUtils; | |
| 33 | import org.apache.commons.lang3.exception.ExceptionUtils; | |
| 34 | import org.apache.tika.mime.MimeTypeException; | |
| 35 | import org.apache.tika.mime.MimeTypes; | |
| 36 | import org.slf4j.Logger; | |
| 37 | import org.w3c.dom.Attr; | |
| 38 | import org.w3c.dom.Comment; | |
| 39 | import org.w3c.dom.Document; | |
| 40 | import org.w3c.dom.Element; | |
| 41 | import org.xwiki.bridge.DocumentAccessBridge; | |
| 42 | import org.xwiki.component.annotation.Component; | |
| 43 | import org.xwiki.model.reference.AttachmentReference; | |
| 44 | import org.xwiki.model.reference.DocumentReference; | |
| 45 | import org.xwiki.model.reference.DocumentReferenceResolver; | |
| 46 | import org.xwiki.rendering.listener.reference.ResourceReference; | |
| 47 | import org.xwiki.rendering.listener.reference.ResourceType; | |
| 48 | import org.xwiki.rendering.renderer.reference.ResourceReferenceSerializer; | |
| 49 | import org.xwiki.xml.XMLUtils; | |
| 50 | import org.xwiki.xml.html.filter.AbstractHTMLFilter; | |
| 51 | ||
| 52 | import com.github.ooxi.jdatauri.DataUri; | |
| 53 | ||
| 54 | /** | |
| 55 | * This filter performs the following transformations on the {@code <img>} tags: | |
| 56 | * <ul> | |
| 57 | * <li>Changes the image source to point to the attached file and adds the XHTML markers (comments) required in order to | |
| 58 | * convert the XHTML to the right wiki syntax. For this you need to specify the "targetDocument" cleaning | |
| 59 | * parameter.</li> | |
| 60 | * <li>Collects the images embedded through the Data URI scheme when the "attachEmbeddedImages" cleaning parameter is | |
| 61 | * set to true. The result can be accessed from the user data associated with the filtered document, under the | |
| 62 | * "embeddedImages" key.</li> | |
| 63 | * <li>Removes the "align" attribute as it can cause problems. First, the office server has a problem with center | |
| 64 | * aligning images (it aligns them to left). Then, the office server uses {@code <br clear"xxx">} to avoid content | |
| 65 | * wrapping around images which is not valid XHTML.</li> | |
| 66 | * </ul> | |
| 67 | * | |
| 68 | * @version $Id: cddbdba11f4af7911c719bebd1bcec5783c66712 $ | |
| 69 | * @since 1.8M1 | |
| 70 | */ | |
| 71 | @Component | |
| 72 | @Named("officeimporter/image") | |
| 73 | @Singleton | |
| 74 | public class ImageFilter extends AbstractHTMLFilter | |
| 75 | { | |
| 76 | private static final String UTF_8 = "UTF-8"; | |
| 77 | ||
| 78 | private static final String EMBEDDED_IMAGES = "embeddedImages"; | |
| 79 | ||
| 80 | @Inject | |
| 81 | private Logger logger; | |
| 82 | ||
| 83 | /** | |
| 84 | * The {@link DocumentAccessBridge} component. | |
| 85 | */ | |
| 86 | @Inject | |
| 87 | private DocumentAccessBridge documentAccessBridge; | |
| 88 | ||
| 89 | /** | |
| 90 | * Used to serialize the image reference as XHTML comment. | |
| 91 | */ | |
| 92 | @Inject | |
| 93 | @Named("xhtmlmarker") | |
| 94 | private ResourceReferenceSerializer xhtmlMarkerSerializer; | |
| 95 | ||
| 96 | /** | |
| 97 | * The component used to parse string document references. | |
| 98 | */ | |
| 99 | @Inject | |
| 100 | @Named("currentmixed") | |
| 101 | private DocumentReferenceResolver<String> stringDocumentReferenceResolver; | |
| 102 | ||
| 103 | 71 |  @Override |
| 104 | public void filter(Document htmlDocument, Map<String, String> cleaningParams) | |
| 105 | { | |
| 106 | 71 | String targetDocumentName = cleaningParams.get("targetDocument"); |
| 107 | 71 | DocumentReference targetDocumentReference = |
| 108 | 71 | targetDocumentName == null ? null : this.stringDocumentReferenceResolver.resolve(targetDocumentName); |
| 109 | ||
| 110 | 71 | boolean attachEmbeddedImages = Boolean.valueOf(cleaningParams.get("attachEmbeddedImages")); |
| 111 | 71 | if (attachEmbeddedImages) { |
| 112 | 2 | htmlDocument.setUserData(EMBEDDED_IMAGES, new HashMap<String, byte[]>(), null); |
| 113 | } | |
| 114 | ||
| 115 | 71 | List<Element> images = filterDescendants(htmlDocument.getDocumentElement(), new String[] {TAG_IMG}); |
| 116 | 71 | for (Element image : images) { |
| 117 | 6 | Attr source = image.getAttributeNode(ATTRIBUTE_SRC); |
| 118 | 6 | if (source != null && targetDocumentReference != null) { |
| 119 | 5 | filterImageSource(source, targetDocumentReference); |
| 120 | } | |
| 121 | ||
| 122 | // The 'align' attribute of images creates a lot of problems. First,the office server has a problem with | |
| 123 | // center aligning images (it aligns them to left). Next, the office server uses <br clear"xxx"> for | |
| 124 | // avoiding content wrapping around images which is not valid XHTML. There for, to be consistent and simple | |
| 125 | // we will remove the 'align' attribute of all the images so that they are all left aligned. | |
| 126 | 6 | image.removeAttribute(ATTRIBUTE_ALIGN); |
| 127 | } | |
| 128 | } | |
| 129 | ||
| 130 | 5 | private void filterImageSource(Attr source, DocumentReference targetDocumentReference) |
| 131 | { | |
| 132 | 5 | String fileName = null; |
| 133 | 5 | try { |
| 134 | 5 | fileName = getFileName(source); |
| 135 | } catch (Exception e) { | |
| 136 | 0 | this.logger.warn("Failed to extract the image file name. Root cause is [{}]", |
| 137 | ExceptionUtils.getRootCauseMessage(e)); | |
| 138 | 0 | this.logger.debug("Full stacktrace is: ", e); |
| 139 | } | |
| 140 | 5 | if (StringUtils.isEmpty(fileName)) { |
| 141 | 2 | return; |
| 142 | } | |
| 143 | ||
| 144 | // Set image source attribute relative to the reference document. | |
| 145 | 3 | AttachmentReference attachmentReference = new AttachmentReference(fileName, targetDocumentReference); |
| 146 | 3 | source.setValue(this.documentAccessBridge.getAttachmentURL(attachmentReference, false)); |
| 147 | ||
| 148 | 3 | ResourceReference imageReference = new ResourceReference(fileName, ResourceType.ATTACHMENT); |
| 149 | 3 | imageReference.setTyped(false); |
| 150 | 3 | Comment beforeComment = source.getOwnerDocument().createComment( |
| 151 | XMLUtils.escapeXMLComment("startimage:" + this.xhtmlMarkerSerializer.serialize(imageReference))); | |
| 152 | 3 | Comment afterComment = source.getOwnerDocument().createComment("stopimage"); |
| 153 | 3 | Element image = source.getOwnerElement(); |
| 154 | 3 | image.getParentNode().insertBefore(beforeComment, image); |
| 155 | 3 | image.getParentNode().insertBefore(afterComment, image.getNextSibling()); |
| 156 | } | |
| 157 | ||
| 158 | 5 | private String getFileName(Attr source) throws MimeTypeException |
| 159 | { | |
| 160 | 5 | String value = source.getValue(); |
| 161 | 5 | String fileName = null; |
| 162 | 5 | @SuppressWarnings("unchecked") |
| 163 | Map<String, byte[]> embeddedImages = | |
| 164 | (Map<String, byte[]>) source.getOwnerDocument().getUserData(EMBEDDED_IMAGES); | |
| 165 | 5 | if (embeddedImages != null && value.startsWith("data:")) { |
| 166 | // An image embedded using the Data URI scheme. | |
| 167 | 2 | DataUri dataURI = DataUri.parse(value, Charset.forName(UTF_8)); |
| 168 | 2 | fileName = dataURI.getFilename(); |
| 169 | 2 | if (StringUtils.isEmpty(fileName)) { |
| 170 | 1 | fileName = String.valueOf(Math.abs(dataURI.hashCode())); |
| 171 | 1 | if (!StringUtils.isEmpty(dataURI.getMime())) { |
| 172 | 1 | String extension = MimeTypes.getDefaultMimeTypes().forName(dataURI.getMime()).getExtension(); |
| 173 | 1 | fileName += extension; |
| 174 | } | |
| 175 | } | |
| 176 | 2 | embeddedImages.put(fileName, dataURI.getData()); |
| 177 | 3 | } else if (!value.contains("://")) { |
| 178 | // A relative path. | |
| 179 | 1 | int separator = value.lastIndexOf('/'); |
| 180 | 1 | fileName = separator < 0 ? value : value.substring(separator + 1); |
| 181 | 1 | try { |
| 182 | // We have to decode the image file name in case it contains URL special characters. | |
| 183 | 1 | fileName = URLDecoder.decode(fileName, UTF_8); |
| 184 | } catch (Exception e) { | |
| 185 | // This shouldn't happen. Use the encoded image file name. | |
| 186 | } | |
| 187 | } | |
| 188 | 5 | return fileName; |
| 189 | } | |
| 190 | } |