<%@ page language="java" contentType="text/html; charset=UTF-8" pageEncoding="UTF-8"%>
<%@ page trimDirectiveWhitespaces="true" %>
<%@ page import="org.crosswire.utils.Utils" %>
<%@ page import="org.crosswire.utils.Sessions" %>
<%@ page import="org.crosswire.xml.XMLBlock" %>
<%@ page import="org.crosswire.xml.XMLTag" %>
<%@ page import="org.crosswire.repo.VersionedRepo" %>
<%@ page import="org.crosswire.utils.HTTPUtils" %>
<%@ page import="org.crosswire.sword.keys.VerseKey" %>
<%@ page import="org.crosswire.sword.keys.ListKey" %>
<%@ page import="org.crosswire.community.projects.ntmss.data.Transcription" %>
<%@ page import="org.crosswire.community.projects.ntmss.data.Document" %>
<%@ page import="org.crosswire.community.projects.ntmss.data.Page" %>
<%@ page import="org.crosswire.community.projects.ntmss.data.Transcription.Convert" %>
<%@ page import="org.crosswire.community.projects.ntmss.data.Transcription.Convert.VerseTranscription" %>
<%@ page import="java.util.Map" %>
<%@ page import="java.util.Vector" %>
<%@ page import="org.apache.log4j.Logger" %>
<%@ page import="org.apache.commons.fileupload.*" %>
<%@ page import="java.util.List" %>
<%
Logger logger = Logger.getLogger("transcript/splitpages");
try {
//dumpCallInfo(request, logger);
String sessionHash = Sessions.getSessionHash(request);
String transcript = request.getParameter("text");
String serverPath = request.getParameter("serverPath");
boolean plainText = "true".equals(request.getParameter("plainText"));
boolean symbolGreek = "true".equals(request.getParameter("symbolGreek"));
boolean ignoreUnderdot = "true".equals(request.getParameter("ignoreUnderdot"));
String format = request.getParameter("format");
int docID = -1; try { docID = Integer.parseInt(request.getParameter("docID")); } catch (Exception e) {}
if (transcript != null && !symbolGreek) { transcript = Transcription.assureUnicode(transcript); }
// See if we're receiving an upload of a transcription
if (transcript == null) {
try {
DiskFileUpload fu = new DiskFileUpload();
// maximum size before a FileUploadException will be thrown
fu.setSizeMax(99000000);
// maximum size that will be stored in memory
fu.setSizeThreshold(4096);
// the location for saving data that is larger than getSizeThreshold()
fu.setRepositoryPath(session.getServletContext().getRealPath("tmp/"));
List fileItems = fu.parseRequest(request);
//logger.info("transcript is null, checking file upload. fileItems, size: " + (fileItems != null ? fileItems.size() : 0));
for (int i = 0; i < fileItems.size(); ++i) {
FileItem fi = (FileItem)fileItems.get(i);
if (fi.isFormField()) {
//logger.info("processing form field: " + fi.getFieldName() + " = " +fi.getString());
if ("docID".equals(fi.getFieldName())) { try { docID = Integer.parseInt(fi.getString()); } catch (Exception e) {} }
if ("plainText".equals(fi.getFieldName())) { plainText = "true".equals(fi.getString()); }
if ("symbolGreek".equals(fi.getFieldName())) { symbolGreek = "true".equals(fi.getString()); }
if ("ignoreUnderdot".equals(fi.getFieldName())) { ignoreUnderdot = "true".equals(fi.getString()); }
if ("format".equals(fi.getFieldName())) { format = fi.getString(); }
}
else {
byte uploadBytes[] = fi.get();
transcript = new String(uploadBytes, symbolGreek ? "iso8859-1" : "UTF-8");
//logger.info("processing upload data, size: " + transcript.length());
}
}
} catch (Exception e) {
logger.debug(e);
}
}
//logger.info("transcript/splitpages(serverPath: " + serverPath + "; plainText: " + plainText + "; symbolGreek: " + symbolGreek + "; ignoreUnderdot: " + ignoreUnderdot + "; format: " + format + "; transcript, size: " + (transcript != null ? transcript.length() : -1));
Transcription transcription = null;
// if we have a server path specified, then we will load our transcription from the server path
//
if (serverPath != null && serverPath.startsWith("/data/")) {
// if we are already XML, then just read the file from the server into the transcript buffer
// as if it was passed directly to us
if (!plainText) {
transcript = Transcription.readFile(serverPath);
}
// otherwise we need to do more processing and we have a Transcription static to do all of that
else transcription = Transcription.fromPlainTextFile(serverPath, ignoreUnderdot, !symbolGreek);
}
// if we have a transcript buffer, we need to be sure it has a header.
if (transcript != null && !symbolGreek && !transcript.trim().startsWith("" + transcript;
//logger.info("transcript/splitpages: transcript.length(): " + (transcript != null ? transcript.length() : 0));
// if we have a transcript buffer, let's construct our transcription from the buffer now
if (transcript != null && transcription == null) {
if (!plainText) {
transcription = new Transcription(transcript);
}
else transcription = Transcription.fromPlainText(transcript, ignoreUnderdot, !symbolGreek);
}
if (transcription != null) {
//logger.info("transcript/splitpages: transcription.getPages().size(): " + transcription.getPages().size());
response.setContentType("text/xml");
%>
<%=HTTPUtils.canonize(transcription.getHead())%>
<%
boolean first = true;
for (String pg : transcription.getPages()) {
// pg = Transcription.alignV11n(pg);
// get ab tags to build verses
StringBuffer abs = new StringBuffer();
for (int i = pg.indexOf(" -1; i = pg.indexOf("", i);
if (e > -1) {
String tag = pg.substring(i+1, e);
String ref = Transcription.getAttribute(tag, "osisID");
if (ref == null || ref.length() < 1) ref = Transcription.getAttribute(tag, "n");
if (ref != null && ref.length() > 0) {
if (abs.length() > 0) abs.append(";");
abs.append(ref);
}
}
else break;
}
// get folio info
String folio = null;
String pageNum = null;
long pageID = -1;
int i = pg.indexOf("", i);
if (e > -1) {
String tag = pg.substring(i+1, e);
folio = Transcription.getAttribute(tag, "n");
try { pageID = Long.parseLong(Transcription.getAttribute(tag, "xml:id")); } catch(Exception e2) {}
String fType = Transcription.getAttribute(tag, "type");
String osisID = Transcription.getAttribute(tag, "osisID");
if (osisID != null && osisID.length() > 0) {
abs = new StringBuffer(osisID);
}
String pageIDAttr = Transcription.getAttribute(tag, "pageID");
if (pageIDAttr != null && pageIDAttr.length() > 0) {
try { pageID = Long.parseLong(pageIDAttr); } catch (Exception e2) {}
}
if ("page".equals(fType)) {
pageNum = folio; folio = null;
}
}
//logger.info("processed page: folio: " + folio + "; page: " + pageNum + "; verses: " + abs);
if (first) {
//logger.info("first page text: " + pg);
first = false;
}
%>
<%=pageNum != null ? "page=\""+pageNum+"\"" : "folio=\""+folio+"\""%>><%= "pagetext".equals(format) ? HTTPUtils.canonize(pg.toString()) : pg %>
<%
}
%>
<%=HTTPUtils.canonize(transcription.getTail())%>
<%
return;
}
} catch (Exception e) { e.printStackTrace(); return; }
%>
<%!
public static void dumpCallInfo(HttpServletRequest request, Logger logger) {
logger.info("transcript/splitpages called...");
logger.info("dumping headers.................................................");
logger.info("request.getContentType: " + request.getContentType());
logger.info("Headers: ");
for (Object o: java.util.Collections.list(request.getHeaderNames())) {
logger.info(o + "=" + request.getHeader(o.toString()));
}
logger.info("Attributes: ");
for (Object o: java.util.Collections.list(request.getAttributeNames())) {
logger.info(o + "=" + request.getAttribute(o.toString()));
}
logger.info("Parameters: ");
for (Object o: java.util.Collections.list(request.getParameterNames())) {
for (String v: request.getParameterValues(o.toString())) {
logger.info(o + "=" + v);
}
}
logger.info("end of dumping headers..........................................");
}
%>
transcript/splitpages
Split transcription XML into pages
Parameters
| text | text of transcription |
| docID | docID, (optional) to help with context |
| format | (xml), pagetext |
| plainText | (false) or true, whether the specified input is plaintext or not |
| symbolGreek | (false) or true, whether the specified input is symbol greek instead of Unicode |
| ignoreUnderdot | (false) or true, whether or not to specially process the underdot as unclear |