List of usage examples for org.apache.poi.hsmf MAPIMessage getHeaders
public String[] getHeaders() throws ChunkNotFoundException
From source file:com.openkm.util.MailUtils.java
License:Open Source License
/** * Convert Outlook Message to Mail/*from w ww . ja va2 s. c om*/ */ public static Mail messageToMail(MAPIMessage msg) throws MessagingException, IOException { com.openkm.bean.Mail mail = new com.openkm.bean.Mail(); Calendar receivedDate = Calendar.getInstance(); Calendar sentDate = Calendar.getInstance(); try { // Can be void if (msg.getMessageDate() != null) { receivedDate.setTime(msg.getMessageDate().getTime()); } // Can be void if (msg.getMessageDate() != null) { sentDate.setTime(msg.getMessageDate().getTime()); } if (msg.getRtfBody() != null) { try { // JEditorPaneRTF2HTMLConverter converter = new JEditorPaneRTF2HTMLConverter(); // mail.setContent(converter.rtf2html(msg.getBodyRTF())); ByteArrayInputStream bais = new ByteArrayInputStream(msg.getRtfBody().getBytes()); ByteArrayOutputStream baos = new ByteArrayOutputStream(); DocConverter.getInstance().rtf2html(bais, baos); mail.setMimeType(MimeTypeConfig.MIME_HTML); mail.setContent(baos.toString().replace("<BR>", "")); IOUtils.closeQuietly(bais); IOUtils.closeQuietly(baos); } catch (Exception e) { throw new MessagingException(e.getMessage(), e); } } else if (msg.getHtmlBody() != null) { mail.setMimeType(MimeTypeConfig.MIME_HTML); mail.setContent(msg.getHtmlBody()); } else if (msg.getTextBody() != null) { mail.setMimeType(MimeTypeConfig.MIME_TEXT); mail.setContent(msg.getTextBody()); } else { mail.setMimeType(MimeTypeConfig.MIME_UNDEFINED); } if (msg.getDisplayTo() != null) { mail.setTo(recipientToString(msg.getDisplayTo())); } else { mail.setTo(new String[] {}); } StringBuilder sb = new StringBuilder(); for (String header : msg.getHeaders()) { sb.append(header).append("\n"); } // Need to replace 0x00 because PostgreSQL does not accept string containing 0x00 // Need to remove Unicode surrogate because of MySQL => SQL Error: 1366, SQLState: HY000 String subject = FormatUtil.trimUnicodeSurrogates(FormatUtil.fixUTF8(msg.getSubject())); mail.setSize(mail.getContent().length()); mail.setSubject((subject == null || subject.isEmpty()) ? NO_SUBJECT : subject); mail.setFrom(msg.getDisplayFrom()); mail.setCc(recipientToString(msg.getDisplayCC())); mail.setBcc(recipientToString(msg.getDisplayBCC())); mail.setReceivedDate(receivedDate); mail.setSentDate(sentDate); } catch (ChunkNotFoundException e) { throw new MessagingException(e.getMessage(), e); } return mail; }
From source file:fr.gouv.culture.vitam.eml.MsgExtract2.java
License:Open Source License
private static String extractInfoSubEmail(MAPIMessage msg, File curDir, Element root, VitamArgument argument, ConfigLoader config) {//w w w .ja v a2 s .c o m File curPath = null; Element keywords = XmlDom.factory.createElement(EMAIL_FIELDS.keywords.name); Element metadata = XmlDom.factory.createElement(EMAIL_FIELDS.metadata.name); String id = config.addRankId(root); curPath = new File(curDir, "MSG_" + id); //System.out.println("start of "+id); String[] values = new String[Keywords.values().length]; for (int i = 0; i < Keywords.values().length; i++) { values[i] = null; } String[] test = null; try { test = msg.getHeaders(); } catch (ChunkNotFoundException e4) { // TODO Auto-generated catch block e4.printStackTrace(); } int lastRank = -1; for (String string : test) { if (string.startsWith(Keywords.NextOne.name) && lastRank >= 0) { String recv = string.substring(Keywords.NextOne.name.length()); if (values[lastRank] == null) { values[lastRank] = recv; } else { values[lastRank] += (Keywords.Received.ordinal() == lastRank ? "\n" : " ") + recv; } } else { if (string.startsWith(Keywords.Date.name)) { values[Keywords.Date.ordinal()] = string.substring(Keywords.Date.name.length()); lastRank = -1; } else if (string.startsWith(Keywords.XOriginalArrivalTime.name)) { values[Keywords.XOriginalArrivalTime.ordinal()] = string .substring(Keywords.XOriginalArrivalTime.name.length()); int pos = values[Keywords.XOriginalArrivalTime.ordinal()].indexOf(" FILETIME="); if (pos > 0) { values[Keywords.XOriginalArrivalTime .ordinal()] = values[Keywords.XOriginalArrivalTime.ordinal()].substring(0, pos); } lastRank = -1; } else if (string.startsWith(Keywords.MessageId.name)) { values[Keywords.MessageId.ordinal()] = string.substring(Keywords.MessageId.name.length()); values[Keywords.MessageId.ordinal()] = StringUtils .removeChevron( StringUtils.unescapeHTML(values[Keywords.MessageId.ordinal()], true, false)) .trim(); lastRank = -1; } else if (string.startsWith(Keywords.InReplyTo.name)) { String reply = StringUtils.removeChevron(StringUtils .unescapeHTML(string.substring(Keywords.InReplyTo.name.length()), true, false)); if (values[Keywords.InReplyTo.ordinal()] == null) { values[Keywords.InReplyTo.ordinal()] = reply; } else { values[Keywords.InReplyTo.ordinal()] += " " + reply; } lastRank = Keywords.InReplyTo.ordinal(); } else if (string.startsWith(Keywords.Received.name)) { String recv = string.substring(Keywords.Received.name.length()); if (values[Keywords.Received.ordinal()] == null) { values[Keywords.Received.ordinal()] = recv; } else { values[Keywords.Received.ordinal()] += "\n" + recv; } lastRank = Keywords.Received.ordinal(); } else if (string.startsWith(Keywords.From.name)) { values[Keywords.From.ordinal()] = string.substring(Keywords.From.name.length()); lastRank = -1; } else if (string.startsWith(Keywords.To.name)) { if (values[Keywords.To.ordinal()] == null) { values[Keywords.To.ordinal()] = string.substring(Keywords.To.name.length()); } else { values[Keywords.To.ordinal()] += " " + string.substring(Keywords.To.name.length()); } lastRank = Keywords.To.ordinal(); } else if (string.startsWith(Keywords.Cc.name)) { if (values[Keywords.Cc.ordinal()] == null) { values[Keywords.Cc.ordinal()] = string.substring(Keywords.Cc.name.length()); } else { values[Keywords.Cc.ordinal()] += " " + string.substring(Keywords.Cc.name.length()); } lastRank = Keywords.Cc.ordinal(); } else if (string.startsWith(Keywords.Bcc.name)) { if (values[Keywords.Bcc.ordinal()] == null) { values[Keywords.Bcc.ordinal()] = string.substring(Keywords.Bcc.name.length()); } else { values[Keywords.Bcc.ordinal()] += " " + string.substring(Keywords.Bcc.name.length()); } lastRank = Keywords.Bcc.ordinal(); } else if (string.startsWith(Keywords.ReturnPath.name)) { if (values[Keywords.ReturnPath.ordinal()] == null) { values[Keywords.ReturnPath.ordinal()] = string.substring(Keywords.ReturnPath.name.length()); } else { values[Keywords.ReturnPath.ordinal()] += " " + string.substring(Keywords.ReturnPath.name.length()); } lastRank = Keywords.ReturnPath.ordinal(); } else if (string.startsWith(Keywords.Importance.name)) { values[Keywords.Importance.ordinal()] = string.substring(Keywords.Importance.name.length()); lastRank = -1; } else if (string.startsWith(Keywords.Priority.name)) { values[Keywords.Priority.ordinal()] = string.substring(Keywords.Priority.name.length()); lastRank = -1; } else if (string.startsWith(Keywords.XFolder.name)) { values[Keywords.XFolder.ordinal()] = string.substring(Keywords.XFolder.name.length()); lastRank = -1; } else if (string.startsWith(Keywords.XSDOC.name)) { values[Keywords.XSDOC.ordinal()] = string.substring(Keywords.XSDOC.name.length()); lastRank = -1; } else if (string.startsWith(Keywords.Sensitivity.name)) { values[Keywords.Sensitivity.ordinal()] = string.substring(Keywords.Sensitivity.name.length()); lastRank = -1; } else { lastRank = -1; } } } /*for (int i = 0; i < Keywords.values().length; i++) { System.out.println(Keywords.values()[i].name()+": "+values[i]); }*/ if (values[Keywords.XFolder.ordinal()] != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.folder.name); sub.addAttribute(EMAIL_FIELDS.folderName.name, values[Keywords.XFolder.ordinal()]); metadata.add(sub); } String fromEmail = values[Keywords.From.ordinal()]; if (fromEmail != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.from.name); addAddress(sub, EMAIL_FIELDS.fromUnit.name, fromEmail); String fromEmail2 = values[Keywords.ReturnPath.ordinal()]; if (fromEmail2 != null && !fromEmail.contains(fromEmail2)) { addAddress(sub, EMAIL_FIELDS.fromUnit.name, fromEmail2); } metadata.add(sub); } else { String fromEmail2 = values[Keywords.ReturnPath.ordinal()]; if (fromEmail2 != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.from.name); addAddress(sub, EMAIL_FIELDS.fromUnit.name, fromEmail2); metadata.add(sub); } } fromEmail = values[Keywords.To.ordinal()]; if (fromEmail != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.toRecipients.name); String[] to = fromEmail.split(","); for (String string2 : to) { addAddress(sub, EMAIL_FIELDS.toUnit.name, string2); } metadata.add(sub); } fromEmail = values[Keywords.Cc.ordinal()]; if (fromEmail != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.ccRecipients.name); String[] to = fromEmail.split(","); for (String string2 : to) { addAddress(sub, EMAIL_FIELDS.ccUnit.name, string2); } metadata.add(sub); } fromEmail = values[Keywords.Bcc.ordinal()]; if (fromEmail != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.bccRecipients.name); String[] to = fromEmail.split(","); for (String string2 : to) { addAddress(sub, EMAIL_FIELDS.bccUnit.name, string2); } metadata.add(sub); } String subject = null; try { subject = msg.getSubject(); } catch (ChunkNotFoundException e3) { // TODO Auto-generated catch block e3.printStackTrace(); } if (subject != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.subject.name); sub.setText(StringUtils.unescapeHTML(subject, true, false)); metadata.add(sub); } subject = null; try { subject = msg.getConversationTopic(); } catch (ChunkNotFoundException e3) { //System.err.println(e3.getMessage()); } if (subject != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.conversationTopic.name); sub.setText(StringUtils.unescapeHTML(subject, true, false)); metadata.add(sub); } if (values[Keywords.Date.ordinal()] != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.sentDate.name); sub.setText(values[Keywords.Date.ordinal()]); metadata.add(sub); } if (values[Keywords.XOriginalArrivalTime.ordinal()] != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.receivedDate.name); sub.setText(values[Keywords.XOriginalArrivalTime.ordinal()]); metadata.add(sub); } if (values[Keywords.Received.ordinal()] != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.receptionTrace.name); String[] traces = values[Keywords.Received.ordinal()].split("\n"); for (String string : traces) { Element sub3 = XmlDom.factory.createElement(EMAIL_FIELDS.trace.name); sub3.setText(string); sub.add(sub3); } metadata.add(sub); } if (values[Keywords.XSDOC.ordinal()] != null) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.emailSize.name); sub.setText(values[Keywords.XSDOC.ordinal()]); metadata.add(sub); } String messageId = values[Keywords.MessageId.ordinal()]; if (messageId != null) { messageId = StringUtils.removeChevron(StringUtils.unescapeHTML(messageId, true, false)).trim(); if (messageId.length() > 1) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.messageId.name); sub.setText(messageId); metadata.add(sub); } } String InReplyToId = values[Keywords.InReplyTo.ordinal()]; if (InReplyToId != null) { InReplyToId = StringUtils.removeChevron(StringUtils.unescapeHTML(InReplyToId, true, false)).trim(); if (InReplyToId.length() > 1) { Element sub = XmlDom.factory.createElement(EMAIL_FIELDS.inReplyTo.name); sub.setText(InReplyToId); if (messageId != null && messageId.length() > 1) { String old = EmlExtract.filEmls.get(InReplyToId); if (old == null) { old = messageId; } else { old += "," + messageId; } EmlExtract.filEmls.put(InReplyToId, old); } metadata.add(sub); } InReplyToId = null; } Element prop = XmlDom.factory.createElement(EMAIL_FIELDS.properties.name); String imp = values[Keywords.Importance.ordinal()]; if (imp != null && imp.length() > 0) { try { int Priority = Integer.parseInt(imp); switch (Priority) { case 5: imp = "LOWEST"; break; case 4: imp = "LOW"; break; case 3: imp = "NORMAL"; break; case 2: imp = "HIGH"; break; case 1: imp = "HIGHEST"; break; default: imp = "LEV" + Priority; } } catch (NumberFormatException e) { // ignore since imp will be used as returned } prop.addAttribute(EMAIL_FIELDS.importance.name, imp); } imp = values[Keywords.Priority.ordinal()]; if (imp != null && imp.length() > 0) { try { int Priority = Integer.parseInt(imp); switch (Priority) { case 5: imp = "LOWEST"; break; case 4: imp = "LOW"; break; case 3: imp = "NORMAL"; break; case 2: imp = "HIGH"; break; case 1: imp = "HIGHEST"; break; default: imp = "LEV" + Priority; } } catch (NumberFormatException e) { // ignore since imp will be used as returned } prop.addAttribute(EMAIL_FIELDS.priority.name, imp); } if (values[Keywords.Sensitivity.ordinal()] != null) { prop.addAttribute(EMAIL_FIELDS.sensitivity.name, values[Keywords.Sensitivity.ordinal()]); } AttachmentChunks[] files = msg.getAttachmentFiles(); boolean Attachments = (files != null && files.length > 0); prop.addAttribute(EMAIL_FIELDS.hasAttachment.name, Boolean.toString(Attachments)); metadata.add(prop); String result = ""; Element identification = null; if (Attachments) { File oldPath = curPath; if (config.extractFile) { File newDir = new File(curPath, id); newDir.mkdir(); curPath = newDir; } identification = XmlDom.factory.createElement(EMAIL_FIELDS.attachments.name); // get the number of attachments for this message int NumberOfAttachments = files.length; identification.addAttribute(EMAIL_FIELDS.attNumber.name, Integer.toString(NumberOfAttachments)); // get a specific attachment from this email. for (int attachmentNumber = 0; attachmentNumber < NumberOfAttachments; attachmentNumber++) { AttachmentChunks attachment = files[attachmentNumber]; if (argument.extractKeyword) { result += " " + extractInfoAttachment(attachment, identification, argument, config, curPath); } else { extractInfoAttachment(attachment, identification, argument, config, curPath); } } curPath = oldPath; } // Plain text e-mail body String body = ""; if (argument.extractKeyword || config.extractFile) { try { body = msg.getTextBody(); } catch (ChunkNotFoundException e2) { // TODO Auto-generated catch block e2.printStackTrace(); } boolean isTxt = true; boolean isHttp = false; if (body == null || body.isEmpty()) { isTxt = false; try { body = msg.getHtmlBody(); } catch (ChunkNotFoundException e1) { // TODO Auto-generated catch block e1.printStackTrace(); } isHttp = true; if (body == null || body.isEmpty()) { isHttp = false; try { body = msg.getRtfBody(); } catch (ChunkNotFoundException e) { // TODO Auto-generated catch block e.printStackTrace(); } } } if (body != null && !body.isEmpty()) { if (config.extractFile) { // XXX FIXME could saved email from HTML Body (clearer) if possible // use curRank in name, and attachment will be under directory named // add currank in field File newDir = new File(curPath, id); newDir.mkdir(); String filenamebody = messageId; if (filenamebody == null || !filenamebody.isEmpty()) { filenamebody = id; } String html = null; if (isHttp) { html = body; } String rtf = null; if (!isTxt && !isHttp) { rtf = body; } if (isTxt) { FileOutputStream output = null; try { output = new FileOutputStream(new File(newDir, filenamebody + ".txt")); byte[] bb = body.getBytes(); output.write(bb, 0, bb.length); } catch (FileNotFoundException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally { if (output != null) { try { output.close(); } catch (IOException e) { } } } try { html = msg.getHtmlBody(); } catch (ChunkNotFoundException e) { // TODO Auto-generated catch block e.printStackTrace(); } } if (html != null && !html.isEmpty()) { FileOutputStream output = null; try { output = new FileOutputStream(new File(newDir, filenamebody + ".html")); byte[] bb = html.getBytes(); output.write(bb, 0, bb.length); } catch (FileNotFoundException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally { if (output != null) { try { output.close(); } catch (IOException e) { } } } html = null; } if (isTxt || isHttp) { try { rtf = msg.getRtfBody(); } catch (ChunkNotFoundException e) { // TODO Auto-generated catch block e.printStackTrace(); } } if (rtf != null && !rtf.isEmpty()) { FileOutputStream output = null; try { output = new FileOutputStream(new File(newDir, filenamebody + ".rtf")); byte[] bb = rtf.getBytes(); output.write(bb, 0, bb.length); } catch (FileNotFoundException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally { if (output != null) { try { output.close(); } catch (IOException e) { } } } rtf = null; } } } } if (metadata.hasContent()) { root.add(metadata); } if (identification != null && identification.hasContent()) { root.add(identification); } if (argument.extractKeyword) { result = body + " " + result; body = null; ExtractInfo.exportMetadata(keywords, result, "", config, null); if (keywords.hasContent()) { root.add(keywords); } } root.addAttribute(EMAIL_FIELDS.status.name, "ok"); //System.out.println("end of "+id); return result; }
From source file:org.alfresco.repo.content.transform.MSGParser.java
License:Apache License
/** * Process header./*from w w w . j a v a2s. c o m*/ * * @param msg * the msg * @param metadata * the metadata * @param xhtml * the xhtml * @throws Exception * the exception */ private void processHeader(MAPIMessage msg, Metadata metadata, XHTMLContentHandler xhtml) throws Exception { StringChunk subjectChunk = msg.getMainChunks().subjectChunk; if (msg.has7BitEncodingStrings()) { CharsetDetector detector = new CharsetDetector(); detector.setText(subjectChunk.getRawValue()); CharsetMatch detect = detector.detect(); if (detect.getConfidence() >= 20) { subjectChunk.set7BitEncoding(detect.getName()); } } String subject = subjectChunk.getValue(); String from = msg.getDisplayFrom(); metadata.set(DublinCore.CREATOR, from); metadata.set(Metadata.MESSAGE_FROM, from); metadata.set(Metadata.MESSAGE_TO, msg.getDisplayTo()); metadata.set(Metadata.MESSAGE_CC, msg.getDisplayCC()); metadata.set(Metadata.MESSAGE_BCC, msg.getDisplayBCC()); metadata.set(DublinCore.TITLE, subject); metadata.set(DublinCore.SUBJECT, msg.getConversationTopic()); try { for (String recipientAddress : msg.getRecipientEmailAddressList()) { if (recipientAddress != null) metadata.add(Metadata.MESSAGE_RECIPIENT_ADDRESS, recipientAddress); } } catch (ChunkNotFoundException he) { } // Will be fixed in POI 3.7 Final // Date - try two ways to find it // First try via the proper chunk if (msg.getMessageDate() != null) { metadata.set(DublinCore.DATE, msg.getMessageDate().getTime()); metadata.set(Office.CREATION_DATE, msg.getMessageDate().getTime()); metadata.set(Office.SAVE_DATE, msg.getMessageDate().getTime()); } else { try { // Failing that try via the raw headers String[] headers = msg.getHeaders(); if (headers != null && headers.length > 0) { for (String header : headers) { if (header.toLowerCase().startsWith("date:")) { String date = header.substring(header.indexOf(':') + 1).trim(); // See if we can parse it as a normal mail date try { Date d = MboxParser.parseDate(date); metadata.set(DublinCore.DATE, d); metadata.set(Office.CREATION_DATE, d); metadata.set(Office.SAVE_DATE, d); } catch (ParseException e) { // Store it as-is, and hope for the best... metadata.set(DublinCore.DATE, date); metadata.set(Office.CREATION_DATE, date); metadata.set(Office.SAVE_DATE, date); } break; } } } } catch (ChunkNotFoundException he) { // We can't find the date, sorry... } } xhtml.element("h1", subject); // Output the from and to details in text, as you // often want them in text form for searching xhtml.startElement("dl"); if (from != null) { header(xhtml, "From", from); } header(xhtml, "To", msg.getDisplayTo()); header(xhtml, "Cc", msg.getDisplayCC()); header(xhtml, "Bcc", msg.getDisplayBCC()); try { header(xhtml, "Recipients", msg.getRecipientEmailAddress()); } catch (ChunkNotFoundException e) { } List<String> attachmentList = new ArrayList<String>(); // // prepare attachments prepareExtractMultipart(xhtml, message, attachmentList); if (attachmentList.size() > 0) { header(xhtml, "Attachments", attachmentList.toString()); } xhtml.endElement("dl"); }
From source file:org.apache.tika.parser.microsoft.OutlookExtractor.java
License:Apache License
/** * Tries to identify the correct encoding for 7-bit (non-unicode) * strings in the file.// w ww .ja va2s . c om * <p>Many messages store their strings as unicode, which is * nice and easy. Some use one-byte encodings for their * strings, but don't always store the encoding anywhere * helpful in the file.</p> * <p>This method checks for codepage properties, and failing that * looks at the headers for the message, and uses these to * guess the correct encoding for your file.</p> * <p>Bug #49441 has more on why this is needed</p> * <p>This is taken verbatim from POI (TIKA-1238) * as a temporary workaround to prevent unsupported encoding exceptions</p> */ private void guess7BitEncoding(MAPIMessage msg) { Chunks mainChunks = msg.getMainChunks(); //sanity check if (mainChunks == null) { return; } Map<MAPIProperty, List<PropertyValue>> props = mainChunks.getProperties(); if (props != null) { // First choice is a codepage property for (MAPIProperty prop : new MAPIProperty[] { MAPIProperty.MESSAGE_CODEPAGE, MAPIProperty.INTERNET_CPID }) { List<PropertyValue> val = props.get(prop); if (val != null && val.size() > 0) { int codepage = ((PropertyValue.LongPropertyValue) val.get(0)).getValue(); String encoding = null; try { encoding = CodePageUtil.codepageToEncoding(codepage, true); } catch (UnsupportedEncodingException e) { //swallow } if (tryToSet7BitEncoding(msg, encoding)) { return; } } } } // Second choice is a charset on a content type header try { String[] headers = msg.getHeaders(); if (headers != null && headers.length > 0) { // Look for a content type with a charset Pattern p = Pattern.compile("Content-Type:.*?charset=[\"']?([^;'\"]+)[\"']?", Pattern.CASE_INSENSITIVE); for (String header : headers) { if (header.startsWith("Content-Type")) { Matcher m = p.matcher(header); if (m.matches()) { // Found it! Tell all the string chunks String charset = m.group(1); if (tryToSet7BitEncoding(msg, charset)) { return; } } } } } } catch (ChunkNotFoundException e) { } // Nothing suitable in the headers, try HTML // TODO: do we need to replicate this in Tika? If we wind up // parsing the html version of the email, this is duplicative?? // Or do we need to reset the header strings based on the html // meta header if there is no other information? try { String html = msg.getHtmlBody(); if (html != null && html.length() > 0) { Charset charset = null; try { charset = detector.detect(new ByteArrayInputStream(html.getBytes(UTF_8)), EMPTY_METADATA); } catch (IOException e) { //swallow } if (charset != null && tryToSet7BitEncoding(msg, charset.name())) { return; } } } catch (ChunkNotFoundException e) { } //absolute last resort, try charset detector StringChunk text = mainChunks.textBodyChunk; if (text != null) { CharsetDetector detector = new CharsetDetector(); detector.setText(text.getRawValue()); CharsetMatch match = detector.detect(); if (match != null && match.getConfidence() > 35 && tryToSet7BitEncoding(msg, match.getName())) { return; } } }