/src/com/atlassian/uwc/converters/vqwiki/AttachmentConverter.java

package com.atlassian.uwc.converters.vqwiki;



	import java.io.File;

import java.util.HashSet;

	import java.util.Set;

	import java.util.TreeSet;

	import java.util.Vector;

	import java.util.regex.Matcher;

	import java.util.regex.Pattern;



	import org.apache.log4j.Logger;



	import com.atlassian.uwc.converters.BaseConverter;

	import com.atlassian.uwc.ui.ConfluenceSettingsForm;

import com.atlassian.uwc.ui.Page;

//deprecated : import com.atlassian.uwc.ui.UWCForm2;



	/**

	 * New 1.

	 * @author P.Hunter

	 *

	 */

	public class AttachmentConverter extends BaseConverter {

		Logger log = Logger.getLogger(this.getClass()); 

		ConfluenceSettingsForm confSettings = null;



		//CONSTANTS - regex patterns

		protected static final String DOTALL = "(?s)"; // flag to enable dotall mode ("." includes newline in search)



		// - need to work out how to capture attachments in tables::

//		private static Pattern attachPattern = Pattern.compile("attach:([^#|]*)[#]*[\n]");



		//private static Pattern attachPattern = Pattern.compile("attach:(.*)[\n]");

		private static Pattern attachPattern = Pattern.compile("attach:(\\w.*\\w)");

		private static final String FILE_SEP = System.getProperty("file.separator");

		



		public void convert(Page vqwikiPage) {



			log.info("Converting VQwiki Attachments -- starting");



			String pageAsTextString = vqwikiPage.getOriginalText();   // input (original) page text

			String convertedPage = "";  // write page to this output string, after successful conversion



			Matcher vqAttachFinder = attachPattern.matcher( pageAsTextString );	

			StringBuffer sb  = new StringBuffer();

			

			String attachmentDir = this.getAttachmentDirectory();

			//Set<File> pageAttachments = new HashSet<File>();



			// scan the page and create a list of attachments

	        // addAttachmentsToPage(vqwikiPage, attachmentDir);



			log.info("Converting VQwiki Attachments -- about to loop...");

			// for each attachment encountered ...

			while (vqAttachFinder.find()) {

				String attachmentName = vqAttachFinder.group(1);

				log.info("** -> attachment:" + attachmentName );

				// write some code to find the attachments for this page as Files

				File attachmentFile = new File(attachmentDir + FILE_SEP + attachmentName);

				// add the attachments to the page, this queues the files to be attached to the Confluence page upon conversion

				vqAttachFinder.appendReplacement(sb, "[^" + attachmentName + "]");

				//TODO

				vqwikiPage.addAttachment(attachmentFile);

				log.info("** -> attached:" + attachmentName );

			}



			// terminate converted text string and pass back to calling program

			vqAttachFinder.appendTail(sb);

			convertedPage = sb.toString();



			vqwikiPage.setConvertedText(convertedPage);

			log.info("Converting VQwiki Attachments -- complete");



		}

		

	    /**

	     * determines which attachments are sought by the page, and attaches them

	     * @param page object to attach pages to

	     */

	    protected void addAttachmentsToPage(Page page, String attachmentDir) {

	    	//what attachments are we looking for?

	    	log.debug("Finding Attachments...");

	    	Vector soughtAttachments = getSoughtAttachmentNames(page);

	    	if (soughtAttachments == null || soughtAttachments.size() == 0)

	    		 return;

	    	

	    	//get filelist from the attachment directory

	    	log.debug("Examining File Directory");

	        File attachmentPageDir = new File(attachmentDir);

	        File files[] = attachmentPageDir.listFiles();

	        if (files==null) {

	            log.info("no attachment files found in directory: "+attachmentDir);

	            return;

	        }

	        

	        // add sought after attachments. This is a recursive method.

	        log.debug("Attaching files to page");

	        addAttachments(page, files, soughtAttachments);

	    }



		/**

		 * recursively look through the images directory for the desired files

		 * and attach them to the given page

		 * @param page object files will be attached to

		 * @param files Array of a directory's files

		 * @param soughtFilenames filenames we are looking to attach

		 */

		private void addAttachments(Page page, File[] files, Vector soughtFilenames) {

			for (File file : files) {

				//check for existence

	            if (!file.exists()) continue;

	            //check for recursion

	            if (file.isDirectory()) {

	            	File moreFiles[] = file.listFiles();

	            	if (moreFiles == null)

	            		continue;

	            	addAttachments(page, moreFiles, soughtFilenames);

	            	continue;

	            }

	            //existing non-Directory file?

	            String filename = file.getName();

	            //is it a file we want to attach?

	            if (foundFile(soughtFilenames, filename)) {

	            	//attach the file

		            log.debug("adding attachment: " + file.getName());

		            page.addAttachment(file);

	            }

	        }

		}



		/**

		 * checks if the given filename can be found in the given vector.

		 * The search is case insensitive. 

		 * @param soughtFilenames

		 * @param filename

		 * @return true if filename is found to be in soughtFilenames

		 */

		protected boolean foundFile(Vector<String> soughtFilenames, String filename) {

			boolean found = soughtFilenames.contains(filename); 

			if (found) return found;

			// check for case insensitivity 

			Pattern caseInsensitiveFilename = Pattern.compile(filename, Pattern.CASE_INSENSITIVE);

			for (String soughtFile : soughtFilenames) {

				Matcher fileFinder = caseInsensitiveFilename.matcher(soughtFile);

				if (fileFinder.matches()) return true;

			}

			return false;

		}



		/**

		 * Determine which images the page is going to need.

		 * Mediawiki images are not associated with a particular page.

		 * @param page the given page

		 * @return String Vector of image names

		 */

		protected Vector<String> getSoughtAttachmentNames(Page page) {

			Vector<String> names = new Vector<String>();

			Set<String> nameSet = new TreeSet<String>();

			String pageText = page.getOriginalText();

			nameSet = getNamesFromImageSyntax(nameSet, pageText);

			nameSet = getNamesFromLinkSyntax(nameSet, pageText);

			names.addAll(nameSet);

			log.debug("found attachment names: " +names.toString());

			return names;

		}



		protected Set<String> getNamesFromImageSyntax(Set<String> nameSet, String pageText) {

			Pattern image = Pattern.compile("!([^!|]+)(?:\\|[^!]+)?!");

			Matcher imageFinder = image.matcher(pageText);

			while (imageFinder.find()) {

				String name = imageFinder.group(1);

				nameSet.add(name);

			}

			return nameSet;

		}

		

		String linkSyntax = 

			"\\[" +			//opening left bracket

			"(" +			//start group

				"?:" +		//but don't capture it

				"[^^]" +	//not a carat

				"[^|]+" +	//anything but a pipe until

				"\\|" +		//a pipe

			")" +			//end group

			"?" +			//and make that group optional

			"\\^" +			//a carat as a part of the string

			"(" +			//start capturing (group1)

				"[^\\]]+" +	//anything but a right bracket until

			")" +			//end capturing (group1)

			"\\]";			//right bracket

		protected Set<String> getNamesFromLinkSyntax(Set<String> nameSet, String pageText) {

			Pattern image = Pattern.compile(linkSyntax);

			Matcher imageFinder = image.matcher(pageText);

			while (imageFinder.find()) {

				String name = imageFinder.group(1);

				nameSet.add(name);

			}

			return nameSet;

		}

	}