Author: asiri Date: 2008-10-28 14:54:04 +0100 (Tue, 28 Oct 2008) New Revision: 13868 Removed: sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/filter/ Modified: sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/OfficeImporterPlugin.java sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/utils/HtmlFilterUtils.java sandbox/xwiki-plugin-officeimporter/src/test/java/com/xpn/xwiki/ plugin/officeconverter/CleanHTMLTest.java Log: Moved all html filter code into a single utility class called HtmlFilterUtils. I thought of introducing some sort of a filter chain (may be chain of responsibility pattern) but it seemed like an over-kill for this scenario. Modified: sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/ xwiki/plugin/officeimporter/OfficeImporterPlugin.java =================================================================== --- sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/OfficeImporterPlugin.java 2008-10-28 11:33:41 UTC (rev 13867) +++ sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/OfficeImporterPlugin.java 2008-10-28 13:54:04 UTC (rev 13868) @@ -57,14 +57,9 @@ import com.xpn.xwiki.doc.XWikiDocument; import com.xpn.xwiki.plugin.XWikiDefaultPlugin; import com.xpn.xwiki.plugin.XWikiPluginInterface; -import com.xpn.xwiki.plugin.officeimporter.filter.EmptyLinkFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.ImageTagFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.PinLiFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.TagRemoveFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.UnderlineLinkFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.XWikiSyntaxEscapeFilter; -import com.xpn.xwiki.plugin.officeimporter.utils.ImporterException; import com.xpn.xwiki.plugin.officeimporter.utils.DocumentType; +import com.xpn.xwiki.plugin.officeimporter.utils.HtmlFilterUtils; +import com.xpn.xwiki.plugin.officeimporter.utils.ImporterException; import com.xpn.xwiki.web.Utils; /** @@ -471,9 +466,7 @@ HTMLCleaner.ROLE), e); } Document document = htmlCleaner.clean(new StringReader(inputHTML)); - - new UnderlineLinkFilter().filter(document); - + HtmlFilterUtils.filterUnderlinedLinks(document); XMLUtils.stripHTMLEnvelope(document); String cleanedHTML = XMLUtils.toString(document); return cleanedHTML; @@ -499,14 +492,12 @@ HTMLCleaner.ROLE), e); } Document document = htmlCleaner.clean(new StringReader(inputHTML)); - - new TagRemoveFilter().filter(document); - new UnderlineLinkFilter().filter(document); - new XWikiSyntaxEscapeFilter().filter(document); - new ImageTagFilter().filter(document); - new PinLiFilter().filter(document); - new EmptyLinkFilter().filter(document); - + HtmlFilterUtils.filterTags(document, new String[]{"style", "script"}); + HtmlFilterUtils.filterUnderlinedLinks(document); + HtmlFilterUtils.filterSytaxChars(document); + HtmlFilterUtils.filterImageLinks(document); + HtmlFilterUtils.filterParagraphTagsInLineItemTags(document); + HtmlFilterUtils.filterEmptyLinks(document); XMLUtils.stripHTMLEnvelope(document); String cleanedHTML = XMLUtils.toString(document); return cleanedHTML; Modified: sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/ xwiki/plugin/officeimporter/utils/HtmlFilterUtils.java =================================================================== --- sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/utils/HtmlFilterUtils.java 2008-10-28 11:33:41 UTC (rev 13867) +++ sandbox/xwiki-plugin-officeimporter/src/main/java/com/xpn/xwiki/ plugin/officeimporter/utils/HtmlFilterUtils.java 2008-10-28 13:54:04 UTC (rev 13868) @@ -1,12 +1,247 @@ package com.xpn.xwiki.plugin.officeimporter.utils; +import java.util.ArrayList; +import java.util.List; + +import org.w3c.dom.Document; +import org.w3c.dom.Element; +import org.w3c.dom.NamedNodeMap; +import org.w3c.dom.Node; +import org.w3c.dom.NodeList; +import org.w3c.dom.Text; + /** * A utility class containing a suite of filter methods used to manipulate Html documents. * * @version $Id$ * @since 1.7M1 */ -public class HtmlFilterUtils +public abstract class HtmlFilterUtils { + /** + * Characters that need to be escaped when jumping from html to xwiki syntax. + */ + private static final List<String> escapeChars = new ArrayList<String>(); + /** + * Static initializer for escape chars. + */ + static { + escapeChars.add("["); + escapeChars.add("]"); + escapeChars.add("{"); + escapeChars.add("}"); + escapeChars.add("*"); + escapeChars.add("~"); + escapeChars.add("_"); + escapeChars.add("-"); + escapeChars.add("1"); + escapeChars.add("#"); + escapeChars.add("$"); + } + + /** + * Removes empty links from html documents. If the label of the link is empty, simply remove the + * tag as in {@code <a/>} or {@code <a href=""/>}. If the label is not null but the href + * attribute is missing, replace the tag with it's label. Like changing {@code <a>something</a>} + * to {@code something}. + * + * @param document The html document. + */ + public static void filterEmptyLinks(Document document) + { + Element root = document.getDocumentElement(); + NodeList links = root.getElementsByTagName("a"); + for (int i = 0; i < links.getLength(); i++) { + Node link = links.item(i); + if (link.getTextContent() == null || link.getTextContent().trim().equals("")) { + link.getParentNode().removeChild(link); + i--; + continue; + } + + Node hrefAttr = link.getAttributes().getNamedItem("href"); + if (hrefAttr == null || hrefAttr.getTextContent().trim().equals("")) { + NodeList children = link.getChildNodes(); + while (children.getLength() > 0) { + link.getParentNode().insertBefore(children.item(0), link); + } + link.getParentNode().removeChild(link); + i--; + } + } + } + + /** + * Replaces the {@code <img>} tags with corresponding {image} macro elements which are + * recognized by xwiki syntax 1.0. Handles image attributes like src, width, height, alt, align. + * + * @param document The html document. + */ + public static void filterImageLinks(Document document) + { + Element root = document.getDocumentElement(); + NodeList imgs = root.getElementsByTagName("img"); + while (imgs.getLength() > 0) { + Node image = imgs.item(0); + String imageCode = generateImageMacroString(image); + Node parent = image.getParentNode(); + Text newImg = document.createTextNode(imageCode); + parent.replaceChild(newImg, image); + } + } + + /** + * Converts a {@code <img>} element into a xwiki syntax 1.0 {image} macro element. + * + * @param imageLink Node representing the image link. + * @return Converted {image} macro string. + */ + private static String generateImageMacroString(Node imageLink) + { + NamedNodeMap attrs = imageLink.getAttributes(); + if (attrs == null) { + return null; + } + StringBuffer sb = new StringBuffer(); + sb.append("{image:"); + if (attrs.getNamedItem("src") != null) { + String src = attrs.getNamedItem("src").getTextContent(); + sb.append(src); + } + if (attrs.getNamedItem("width") != null) { + String width = attrs.getNamedItem("width").getTextContent(); + sb.append("|width=" + width); + } + if (attrs.getNamedItem("height") != null) { + String height = attrs.getNamedItem("height").getTextContent(); + sb.append("|height=" + height); + } + if (attrs.getNamedItem("alt") != null) { + String alt = attrs.getNamedItem("alt").getTextContent(); + sb.append("|alt=" + alt); + } + if (attrs.getNamedItem("align") != null) { + String align = attrs.getNamedItem("align").getTextContent(); + sb.append("|align=" + align); + } + sb.append("}"); + return sb.toString(); + } + + /** + * Removes the starting {@code <p>} tags found within {@code <li>} tags. This is useful since + * such formations are not properly handled in xwiki 1.0 syntax. + * + * @param document The html document. + */ + public static void filterParagraphTagsInLineItemTags(Document document) + { + Element root = document.getDocumentElement(); + NodeList lists = root.getElementsByTagName("li"); + for (int i = 0; i < lists.getLength(); i++) { + Node list = lists.item(i); + Node firstChild = list.getFirstChild(); + if (firstChild.getNodeName() != null && firstChild.getNodeName().equals("p")) { + NodeList childchildren = firstChild.getChildNodes(); + while (childchildren.getLength() > 0) { + list.insertBefore(childchildren.item(0), firstChild); + } + list.removeChild(firstChild); + } + } + } + + /** + * Removes all listed tags from the given html document. + * + * @param document The html document. + * @param tags Tags to be removed. + */ + public static void filterTags(Document document, String[] tags) + { + Element root = document.getDocumentElement(); + for (String tag : tags) { + NodeList toBeRemovedTags = root.getElementsByTagName(tag); + while (toBeRemovedTags.getLength() > 0) { + Node t = toBeRemovedTags.item(0); + t.getParentNode().removeChild(t); + } + } + } + + /** + * Strips off underline tags surrounding links like {@code <u><a href="something">link</a></u>}. + * + * @param document The html document. + */ + public static void filterUnderlinedLinks(Document document) + { + Element root = document.getDocumentElement(); + NodeList links = root.getElementsByTagName("a"); + for (int i = 0; i < links.getLength(); i++) { + Node link = links.item(i); + Node parent = link.getParentNode(); + String parentName = parent.getNodeName(); + if (parentName != null && (parentName.equals("u") || parentName.equals("del"))) { + parent.getParentNode().insertBefore(link, parent); + parent.getParentNode().removeChild(parent); + } + } + } + + /** + * Escapes the xwiki sytax characters from the given html document. Example : {@code [} will be + * replaced by {@code \]}. + * + * @param document The html document. + */ + public static void filterSytaxChars(Document document) + { + Element root = document.getDocumentElement(); + escapeNode(root); + } + + /** + * Escapes xwiki syntax characters within the given node's content. + * + * @param node The node which is to be examined. + */ + private static void escapeNode(Node node) + { + NodeList nodes = node.getChildNodes(); + for (int i = 0; i < nodes.getLength(); i++) { + Node next = nodes.item(i); + if (next instanceof Text) { + String text = next.getTextContent(); + text = escapeText(text); + next.setTextContent(text); + } else { + if (next.hasChildNodes()) { + escapeNode(next); + } + } + } + } + + /** + * Escapes xwiki syntax characters within the given string. + * + * @param text The string to be examined. + * @return The syntax escaped string. + */ + private static String escapeText(String text) + { + StringBuffer sb = new StringBuffer(); + for (int i = 0; i < text.length(); i++) { + char x = text.charAt(i); + if (escapeChars.contains(String.valueOf(x))) { + sb.append("\\"); + sb.append(String.valueOf(x)); + } else { + sb.append(x); + } + } + return sb.toString(); + } } Modified: sandbox/xwiki-plugin-officeimporter/src/test/java/com/xpn/ xwiki/plugin/officeconverter/CleanHTMLTest.java =================================================================== --- sandbox/xwiki-plugin-officeimporter/src/test/java/com/xpn/xwiki/ plugin/officeconverter/CleanHTMLTest.java 2008-10-28 11:33:41 UTC (rev 13867) +++ sandbox/xwiki-plugin-officeimporter/src/test/java/com/xpn/xwiki/ plugin/officeconverter/CleanHTMLTest.java 2008-10-28 13:54:04 UTC (rev 13868) @@ -27,13 +27,7 @@ import org.xwiki.xml.XMLUtils; import org.xwiki.xml.html.HTMLCleaner; -import com.xpn.xwiki.plugin.officeimporter.filter.EmptyLinkFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.HTMLFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.ImageTagFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.PinLiFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.TagRemoveFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.UnderlineLinkFilter; -import com.xpn.xwiki.plugin.officeimporter.filter.XWikiSyntaxEscapeFilter; +import com.xpn.xwiki.plugin.officeimporter.utils.HtmlFilterUtils; import com.xpn.xwiki.plugin.officeimporter.utils.ImporterException; import com.xpn.xwiki.test.AbstractXWikiComponentTestCase; @@ -121,14 +115,12 @@ private void test(String input, String expected) throws ImporterException { Document document = cleaner.clean(new StringReader(input)); - - new TagRemoveFilter().filter(document); - new UnderlineLinkFilter().filter(document); - new XWikiSyntaxEscapeFilter().filter(document); - new ImageTagFilter().filter(document); - new PinLiFilter().filter(document); - new EmptyLinkFilter().filter(document); - + HtmlFilterUtils.filterTags(document, new String[]{"style", "script"}); + HtmlFilterUtils.filterUnderlinedLinks(document); + HtmlFilterUtils.filterSytaxChars(document); + HtmlFilterUtils.filterImageLinks(document); + HtmlFilterUtils.filterParagraphTagsInLineItemTags(document); + HtmlFilterUtils.filterEmptyLinks(document); XMLUtils.stripHTMLEnvelope(document); String actual = XMLUtils.toString(document); assertEquals(HEAD + expected + FOOT, actual); _______________________________________________ notifications mailing list notifications(a)xwiki.org http://lists.xwiki.org/mailman/listinfo/notifications