package ca.ucalgary.seahawk.util; import org.biomoby.registry.meta.Registry; import org.biomoby.shared.*; import org.biomoby.shared.data.*; import org.biomoby.shared.parser.MobyTags; import javax.xml.parsers.*; import java.io.*; import java.net.*; import java.util.Vector; public class HTMLUtils{ public final static String BINARY_DATA_LABEL = "[binary data not shown]"; public final static String ACCEPTABLE_IMAGE_DATATYPE = "WebFormatImage"; //from the ontology, should eventually be configurable public static String encapsulateBinaryData(URL sourceMobyData, String html) throws Exception{ return encapsulateBinaryData(sourceMobyData, html, null); } public static String encapsulateBinaryData(URL sourceMobyData, String html, Registry registry) throws Exception{ MobyContentInstance dataPackage = MobyDataUtils.fromXMLDocument(sourceMobyData, registry); // Look for binary data and replace it with an image ref (if its an image // displayable in our viewer), or simply "(binary data)". for(MobyDataJob job: dataPackage.values()){ for(MobyDataInstance primary: job.values()){ Iterable values = null; if(primary instanceof MobyDataObjectSet || primary instanceof MobyDataObjectVector){ values = (Iterable) primary; } else if(primary instanceof MobyDataObject){ values = new Vector(); ((Vector) values).add((MobyDataObject) primary); } else{ // ignore secondaries, they can't be binary continue; } for(MobyDataObject data: values){ if(data instanceof MobyDataBytes){ MobyDataBytes binaryMobyObject = (MobyDataBytes) data; //System.err.println("Replacing binary data in display of "+binaryMobyObject.getName()); String encodedData = binaryMobyObject.get(MobyDataBytes.ENCODED_MEMBER_NAME) .getValue().toString(); if(html.indexOf(encodedData) == -1){ System.err.println("Could not find the encoded data in the html given!!!"); } // if an image, save and show it if(binaryMobyObject.getDataType().inheritsFrom(ACCEPTABLE_IMAGE_DATATYPE)){ String imgSuffix = binaryMobyObject.getDataType().getName(); if(imgSuffix.length() < 3){ imgSuffix = ".img"; } else{ imgSuffix = "."+imgSuffix.substring(0, 3).toLowerCase(); } File imageFile = File.createTempFile("seahawkSavedImage", imgSuffix); imageFile.deleteOnExit(); FileOutputStream fos = new FileOutputStream(imageFile); fos.write(binaryMobyObject.getBytes()); fos.close(); html = html.replace(encodedData, "

"); } // Otherwise put a placeholder else{ html = html.replace(encodedData, BINARY_DATA_LABEL); } } } } } return html; } /** * Extracts a URL string from a given piece of text iof that's all it contains. * Especially suited for dropped URL's etc. from browsers, as it cleans up the HTML junk * browsers wrap links in. */ public static String checkForHyperlinkText(String text){ if(text.indexOf("") == text.length()-11){ return text.substring(15, text.indexOf("\"",15)); } else if(text.indexOf("") == text.length()-4){ //unix return text.substring(9, text.indexOf("\"",9)); } else if(text.indexOf("") != -1){ int startHref = text.indexOf(""+text+""; } StringBuffer result = new StringBuffer(""); int lineCharCount = 0; String[] tokens = text.split("[ \t\n]"); for(String word: tokens){ while(word.length()+1 > maxLineLength){ //single word is bigger than preset width, hyphenate int cutPoint = maxLineLength - lineCharCount - 1; if(cutPoint < word.length()-1){ cutPoint = word.length()-1; } if(cutPoint < 0){ break; } result.append(word.substring(0, cutPoint)+"-
"); lineCharCount = 0; word = word.substring(cutPoint); } if(lineCharCount != 0 && lineCharCount + word.length() > maxLineLength){ result.append("
" + word + " "); lineCharCount = word.length()+1; continue; } result.append(word+" "); lineCharCount += word.length()+1; } result.append(""); return result.toString(); } /** * Given an HTML fragment (e.g. cut'n'pasted from a browser), see if we unescape the * HTML we get a MOBY XML block (e.g. from the biomoby.org web site examples). */ public static MobyContentInstance checkForMobyXML(String text){ if(//(text.indexOf("moby:") != -1 || text.indexOf("articleName=\"") != -1) && text.indexOf("namespace=\"") == -1 || text.indexOf("id=\"") == -1){ return null; } MobyContentInstance content = null; String xmltext = text; if(xmltext.indexOf("??") == 0){ xmltext = xmltext.substring(2); } // See if the xml needs unescaping (e.g. if preformatted HTML dropped from Firefox or Thunderbird) int preStart = xmltext.indexOf("")+1); xmltext = xmltext.replaceAll(">", ">"); xmltext = xmltext.replaceAll("<", "<"); xmltext = xmltext.replaceAll("&", "&"); } xmltext = xmltext.trim(); // remove leading and trailing spaces try{ DocumentBuilderFactory dbf = DocumentBuilderFactory.newInstance(); dbf.setNamespaceAware (true); DocumentBuilder db = dbf.newDocumentBuilder(); if(xmltext.indexOf("xmlns:moby") == -1){ // Ensure that the moby: prefix is declared in the first tag xmltext = xmltext.replaceFirst("<([A-Za-z0-9_][^> \t\n]*)", "<$1 xmlns:moby=\""+ MobyPrefixResolver.MOBY_XML_NAMESPACE+"\""); } if(xmltext.indexOf("\n" + xmltext; } org.w3c.dom.Element data = db.parse(new ByteArrayInputStream(xmltext.getBytes())).getDocumentElement(); if(data.getLocalName().equals(MobyTags.MOBY)){ content = MobyDataUtils.fromXMLDocument(data, SeahawkOptions.getRegistry()); } else if(data.getLocalName().equals(MobyTags.MOBYCONTENT)){ content = new MobyContentInstance(data, SeahawkOptions.getRegistry()); } else if(data.getLocalName().equals(MobyTags.MOBYDATA)){ content = new MobyContentInstance(); content.parseDataGroup(data, SeahawkOptions.getRegistry()); } else{ // Any other data can be processed by the base object class content = new MobyContentInstance(MobyDataObject.createInstanceFromDOM(data, SeahawkOptions.getRegistry()), "Drag 'n' Drop data"); } } catch(Exception e){ e.printStackTrace(); System.err.println("This issue may or may not be related to your choice of registry ("+ (SeahawkOptions.getRegistry() == null ? "MOBY default" : SeahawkOptions.getRegistry().getSynonym()) +")"); System.err.println(xmltext); } return content; } /** * Convenience method to grab all data from a URL into a String. */ public static String getURLContents(URL u) throws IOException{ StringBuffer buffer = new StringBuffer(); LineNumberReader reader = new LineNumberReader(new InputStreamReader(u.openStream())); for(String line = reader.readLine(); line != null; line = reader.readLine()){ buffer.append(line+"\n"); } return buffer.toString(); } /** * Convenience method to grab all data from a URL as a byte array (in case it's not text) */ public static byte[] getURLContentBytes(URL u) throws IOException{ InputStream urlStream = u.openStream(); byte[] byteBufferChunk = new byte[1024]; ByteArrayOutputStream byteBuffer = new ByteArrayOutputStream(); for(int r = urlStream.read(byteBufferChunk, 0, 1024); r != -1; r = urlStream.read(byteBufferChunk, 0, 1024)){ byteBuffer.write(byteBufferChunk, 0, r); } return byteBuffer.toByteArray(); } /** * Convenience method to grab all data from a input stream into a String. */ public static String getInputStreamContents(InputStream resourceStream) throws IOException{ StringBuffer xml = new StringBuffer(); // Copy data verbatim 1 k at a time byte[] dataBuffer = new byte[1024]; for(int numBytesRead = resourceStream.read(dataBuffer); numBytesRead != -1; numBytesRead = resourceStream.read(dataBuffer)){ xml.append(new String(dataBuffer, 0, numBytesRead)); } return xml.toString(); } }