/*
 * $Id: tocgen.bsh,v 1.3 2004/01/30 06:46:33 mitch Exp $
 * 
 * OSI Certified Open Source Software (see www.opensource.org for details)
 * Licensed under the BSD license:
 *
 * Copyright (c) 2004, FullSpan Software (www.fullspan.com)
 * All rights reserved
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * - Redistributions of source code must retain the above copyright notice,
 * this list of conditions and the following disclaimer.
 *
 * - Redistributions in binary form must reproduce the above copyright notice,
 * this list of conditions and the following disclaimer in the documentation
 * and/or other materials provided with the distribution. 
 *
 * - Neither the name of FullSpan Software nor the names of its contributors
 * may be used to endorse or promote products derived from this software
 * without specific prior written permission. 
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
 * THE POSSIBILITY OF SUCH DAMAGE.
 */

// Run with beanshell (www.beanshell.org):
//
//   java bsh.Interpreter tocgen.bsh INFILE OUTFILE STYLESHEET
//
// See tocgen.html for detailed usage information.

import java.io.*;
import java.util.*;
import java.util.regex.*;

import javax.xml.transform.*;
import javax.xml.transform.stream.*;

import org.jdom.*;
import org.jdom.filter.*;
import org.jdom.input.*;
import org.jdom.output.*;
import org.jdom.transform.*;
import org.jdom.xpath.*;

// Constants

   final String[]       usageLines =
   {
      "",
      "TOCGen v1.0",
      "Usage:",
      "",
      "  java bsh.Interpreter tocgen.bsh INFILE OUTFILE STYLESHEET",
      "",
      "Adds a Table of Contents to an XHTML document.  See tocgen.html for",
      "detailed usage information.",
      ""
   };

   final boolean        trace = false;

   // Because XHTML input files have a namespace, we need to qualify our
   // XPath expressions.  We just use an arbitrary "x" as the namespace prefix.
   // See:  http://www.servlets.com/archive/servlet/ReadMsg?msgId=425772&listName=jdom-interest
   final Namespace      NS = Namespace.getNamespace("x", "http://www.w3.org/1999/xhtml");

   final int            MAX_HEADING_LEVEL = 6;
   final String         ANCHOR_PREFIX = "toc-";

   // Pattern for heading number
   //    Optional whitespace
   //    Sequence of numbers and dots
   //    Optional whitespace
   final Pattern        HEADING_PATTERN =
      Pattern.compile("^\\s*[\\d\\.]+\\s*");
   final Matcher        HEADING_MATCHER = HEADING_PATTERN.matcher("");

   final Filter         TEXT_AND_ELEMENT_FILTER =
      new ContentFilter(ContentFilter.TEXT | ContentFilter.ELEMENT);

// Objects (closures == pseudo-classes - see BeanShell User's Guide)

   Heading(elem)
   {
      m_elem = elem;
      m_level = getHeadingLevel(m_elem.getName());
      m_headingNumberText = "";
      m_anchorId = "";
      m_contentText = "";

      void setHeadingNumber(headingNumbers, firstLevel)
      {
         m_headingNumberText = getHeadingNumberText(headingNumbers, firstLevel);
         m_anchorId = ANCHOR_PREFIX + m_headingNumberText;
      }

      void setHeadingText()
      {
         textNode = getFirstTextNode(m_elem);

         if (textNode == null)
         {
            throw new RuntimeException("Missing text for heading: " + m_elem);
         }

         text = textNode.getTextTrim();
         
         HEADING_MATCHER.reset(text);
         text = HEADING_MATCHER.replaceFirst("");
         text = m_headingNumberText + " " + text;

         textNode.setText(text);

         // Get the text for the table of contents for this heading.
         // If we just use m_elem.getTextTrim() this will only give us the
         // direct text.
         //
         // For example, if we have a heading like this:
         //
         //  <h3>This is an <i>important</i> chapter</h3>
         //
         // this would give us the contents entry: "This is an chapter",
         // because the italicized text is in a child element.
         // So we need to extract the text recursively from all child
         // elements and use that for the contents entry.  In the
         // above example this would give the contents entry:
         // "This is an important chapter".

         m_contentText = getAllText(m_elem);
      }

      return this;
   }

// Utility methods

   XPath getXPath(path)
   {
      xp = XPath.newInstance(path);
      xp.addNamespace(NS);
      return xp;
   }

   /**
    * Return true if the node is an Element with name h1...h6,
    * otherwise return false
    */
   boolean isHeading(node)
   {
      if (!(node instanceof Element))
      {
         return false;
      }

      name = node.getName();
      
      if (name != null && name.length() == 2 && name.startsWith("h") &&
          name.charAt(1) >= '1' && name.charAt(1) <= '6')
      {
         return true;
      }

      return false;
   }

   /**
    * return 1, 2, ... for elements "h1", "h2", ... 
    */
   int getHeadingLevel(elemName)
   {
      
      return elemName.charAt(1) - '0';
   }

   /**
    * Return the heading number as a dotted string, e.g. "2.1.4"
    */
   String getHeadingNumberText(headingNumbers, firstLevel)
   {
      StringBuffer buf = new StringBuffer();

      for (i = firstLevel; i < headingNumbers.length; i++)
      {
         if (headingNumbers[i] == 0)
         {
            break;
         }

         if (i > firstLevel)
         {
            buf.append(".");
         }

         buf.append(headingNumbers[i]);
      }

      return buf.toString();
   }

   /**
    * Return the first child node of type Text, searching recursively
    * from the given node.  Search is depth-first.
    */
   Text getFirstTextNode(node)
   {
      if (node instanceof Text)
      {
         return node;
      }

      // node instanceof Element

      content = node.getContent(TEXT_AND_ELEMENT_FILTER);

      for (contentNode: content)
      {
         result = getFirstTextNode(contentNode);

         if (result != null)
         {
            return result;
         }
      }

      return null;
   }

   /**
    * Return the text of all nodes at and below the given node, searching
    * recursively.  Search is depth-first.
    */
   String getAllText(node)
   {
      textList = new ArrayList();
      getAllText(node, textList);

      buf = new StringBuffer();

      for (text: textList)
      {
         buf.append(text);
         buf.append(" ");
      }

      return buf.toString();
   }

   /**
    * Find the text of all nodes at and below the given node, searching
    * recursively.  Search is depth-first.  Append the text found to
    * the specified textList.
    */
   void getAllText(node, textList)
   {
      if (node instanceof Text)
      {
         textList.add(node.getTextTrim());
      }
      else // node instanceof Element
      {
         content = node.getContent(TEXT_AND_ELEMENT_FILTER);
   
         for (contentNode: content)
         {
            getAllText(contentNode, textList);
         }
      }
   }



   
// ========== Begin main body

// Process command line

   if (bsh.args.length != 3)
   {
      for (usageLine: usageLines)
      {
         print(usageLine);
      }

      exit();
   }

   inFilename = bsh.args[0];
   inFile = new File(inFilename);

   if (!inFile.exists())
   {
      print("Input file does not exist: " + inFilename);
      exit();
   }

   outFilename = bsh.args[1];
   outFile = new File(outFilename);

   stylesheetFilename = bsh.args[2];
   stylesheetFile = new File(stylesheetFilename);

   if (!stylesheetFile.exists())
   {
      print("Stylesheet file does not exist: " + stylesheetFilename);
      exit();
   }

// Load the input file

   saxBuilder = new SAXBuilder();
   jdomDoc = saxBuilder.build(inFile);
   rootElem = jdomDoc.getRootElement();

// Find where we will insert the toc

   tocElem = getXPath("//x:div[@id='toc']").selectSingleNode(rootElem);
   
   if (tocElem == null)
   {
      print("Missing toc placeholder.  Expected: <div id=\"toc\">");
      exit();
   }

// Get all of the headings

   bodyElem = getXPath("x:body").selectSingleNode(rootElem);
   allNodes = getXPath("//x:*").selectNodes(bodyElem);
   headings = new ArrayList();

   for (node: allNodes)
   {
      if (isHeading(node))
      {
         headings.add(Heading(node));
      }
   }

   if (headings.size() == 0)
   {
      print("There are no headings in the source document.");
      exit();
   }


   if (trace)
   {
      for (heading: headings)
      {
         print(heading.m_elem.getName() + ": " + heading.m_elem.getTextTrim());
      }
   }

   // The headingNumbers array keeps track of the current heading numbers.
   // Index [0] is unused, indexes [1] up to and including [6] represent
   // the current heading number for h1...h6.

   int[] headingNumbers = new int[MAX_HEADING_LEVEL + 1];

   headingNumbers[0] = -1;

   for (i = 1; i <= MAX_HEADING_LEVEL; i++)
   {
      headingNumbers[i] = 0;
   }

   firstLevel = 0;

   for (heading: headings)
   {
      if (firstLevel == 0)
      {
         firstLevel = heading.m_level;
      }
      else if (heading.m_level < firstLevel)
      {
         print("Error: the first heading in the document must have the " +
            "lowest hN number, this heading is not valid: " +
            heading.m_elem.getName() + ": " + heading.m_elem.getTextTrim());
         exit();
      }

      headingNumbers[heading.m_level]++;

      for (i = firstLevel; i < heading.m_level; i++)
      {
         if (headingNumbers[i] <= 0)
         {
            print("Error: gap in heading levels.  This heading is more than " +
               "one level deeper than any previous heading: " +
               heading.m_elem.getName() + ": " + heading.m_elem.getTextTrim());
            exit();
         }
      }

      // When a more prominent heading (with a lower hN number) is found,
      // reset the numbers for less prominent headings (those with
      // higher hN numbers).

      for (i = heading.m_level + 1; i <= MAX_HEADING_LEVEL; i++)
      {
         headingNumbers[i] = 0;
      }

      heading.setHeadingNumber(headingNumbers, firstLevel);
      heading.setHeadingText();

      if (trace)
      {
         print(heading.m_headingNumberText + ": " + heading.m_elem.getTextTrim());
      }
   }

   // Create an in-memory XML document, tocDataDoc, to hold the table of
   // contents data.  For example:
   //
   //  <toc>
   //    <entry refid="toc_1" level="0">1 A Heading</entry>
   //    <entry refid="toc_1.1" level="1">1.1 Another Heading</entry>
   //  </toc>

   tocDataRoot = new Element("toc");
   tocDataDoc = new Document(tocDataRoot);

   for (heading: headings)
   {
      // Remove any existing toc-* anchors (this allows the file to be
      // processed multiple times without creating duplicate tags).

      anchors = getXPath("x:a[@id]").selectNodes(heading.m_elem);

      for (anchor: anchors)
      {
         if (anchor.getAttributeValue("id").startsWith(ANCHOR_PREFIX))
         {
            anchor.detach();
         }
      }

      // Add one new anchor tag to each heading.

      anchor = new Element("a", heading.m_elem.getNamespace()).
         setAttribute("id", heading.m_anchorId);
      heading.m_elem.addContent(anchor);

      // Add this heading to the TOC data document

      tocDataElem = new Element("entry").
         setText(heading.m_contentText).
         setAttribute("refid", heading.m_anchorId);
      tocDataElem.setAttribute("level",
         String.valueOf(heading.m_level - firstLevel));
      tocDataRoot.addContent(tocDataElem);
   }

   // Create TOC display by merging TOC data with stylesheet

   transformer = TransformerFactory.newInstance().
      newTransformer(new StreamSource(stylesheetFile));

   JDOMResult tocResult = new JDOMResult();
   transformer.transform(new JDOMSource(tocDataDoc), tocResult);

   // Delete all the current content in the main doc TOC (if any)

   tocElem.getContent().clear();

   // Merge transformed doc into main doc

   for (resultNode: tocResult.getResult())
   {
      if (resultNode instanceof Element)
      {
         resultNode.setNamespace(tocElem.getNamespace());
      }

      tocElem.addContent(resultNode);
   }

   // Output the merged doc

   outStream = new FileOutputStream(outFile);

   XMLOutputter outputter = new XMLOutputter();
   outputter.setEncoding("UTF-8");
   outputter.setTextTrim(true);
   outputter.setIndent("   ");
   outputter.setNewlines(true);
   outputter.output(jdomDoc, outStream);

   outStream.close();