source: other-projects/hathitrust/wcsa/vol-checker/src/org/hathitrust/extractedfeatures/VolumeCheck.java@ 31279

package org.hathitrust.extractedfeatures;

import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.io.PrintWriter;
import java.util.ArrayList;
import java.util.HashMap;

import javax.servlet.ServletConfig;
import javax.servlet.ServletException;
import javax.servlet.annotation.WebServlet;
import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;

/**
 * Servlet implementation class VolumeCheck
 */
@WebServlet("/VolumeCheck")
public class VolumeCheck extends HttpServlet {
    private static final long serialVersionUID = 1L;

    protected static int HASHMAP_INIT_SIZE = 13800000;
    protected static String DATA_DIR="D:/cygwin64/home/davidb/research/code-managed/hathitrust/wcsa/extracted-features-solr/trunk/solr-ingest";
    protected static HashMap<String,Boolean> id_check_ = null;
    
    
    public VolumeCheck() {
        if (id_check_ == null) {
            id_check_ = new HashMap<String,Boolean>(HASHMAP_INIT_SIZE);
            storeIDs(DATA_DIR + "/full-listing.txt");
        }
    }

    protected void storeIDs(String filename)
    {       
        try {
            FileReader fr = new FileReader(filename);
            BufferedReader br = new BufferedReader(fr);
            
            long line_num = 1;
            String line;
            System.err.println("Loading in volume IDS: " + filename);
            
            String opt_file_ext = ".json.bz2";
            
            while ((line = br.readLine()) != null) {
                
                String file_id = line.substring(line.lastIndexOf("/")+1);
                String id = null;
                if (file_id.endsWith(opt_file_ext)) {
                    id = file_id.substring(0,file_id.lastIndexOf(opt_file_ext));
                }
                else {
                    id = file_id;
                }
                
                id = id.replaceAll("\\+", ":").replaceAll("=", "/");
                         
                id_check_.put(id, true);
                
                if ((line_num % 100000) == 0) {
                    System.err.println("sample id = " + id);
                    System.err.println("Passed line: " + line_num);
                }
                line_num++;
                
            }
            
            br.close();
        }
        catch (Exception e) {
            e.printStackTrace();
        }
        
    }
    /**
     * @see Servlet#init(ServletConfig)
     */
    public void init(ServletConfig config) throws ServletException {
        // currently everything we need to happen at start up can be done in the constructor
    }

    /**
     * @see HttpServlet#doGet(HttpServletRequest request, HttpServletResponse response)
     */
    protected void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {
        PrintWriter pw = response.getWriter();
        
        String cgi_ids = request.getParameter("ids");
        String cgi_id = request.getParameter("id");
        
        if (cgi_ids != null) {
            response.setContentType("application/json");
    
            String[] ids = cgi_ids.split(",");
            int ids_len = ids.length;
            
            pw.append("{");
            
            for (int i=0; i<ids_len; i++) {
                String id = ids[i];
                
                Boolean exists = id_check_.get(id);
                boolean status = (exists != null) ? exists : false;
                
                if (i>0) {
                    pw.append(",");
                }
                pw.append("\"" + id + "\":" + status );
            }
            pw.append("}");
            
        }
        else if (cgi_id != null) {
            response.setContentType("application/json");
            
            String id = cgi_id;
            Boolean exists = id_check_.get(id);
            boolean status = (exists != null) ? exists : false;
            pw.append("{'" + id + "':" + status + "}");
        }
        else {

            pw.append("General Info: Number of HTRC Volumes in check-list = " + id_check_.size());
                        
        }
        //pw.close();
        
    }

    /**
     * @see HttpServlet#doPost(HttpServletRequest request, HttpServletResponse response)
     */
    protected void doPost(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {
        doGet(request, response);
    }

}