source: other-projects/hathitrust/wcsa/extracted-features-solr/trunk/solr-ingest/src/main/java/org/hathitrust/extractedfeatures/ProcessForWhitelist.java@ 31250

package org.hathitrust.extractedfeatures;

import java.io.BufferedInputStream;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.Serializable;
import org.apache.commons.cli.*;

import org.apache.spark.api.java.*;
import org.apache.spark.api.java.function.Function2;
import org.apache.spark.api.java.function.PairFunction;
import org.apache.spark.util.DoubleAccumulator;
import scala.Tuple2;

import org.apache.spark.SparkConf;

public class ProcessForWhitelist implements Serializable
{
    private static final long serialVersionUID = 1L;

    // Following details on number of partitions to use given in 
    //  "Parallelized collections" section of:
    //   https://spark.apache.org/docs/2.0.1/programming-guide.html
    //
    // For a more detailed discussion see:
    //   http://blog.cloudera.com/blog/2015/03/how-to-tune-your-apache-spark-jobs-part-2/
    
    protected static final int DEFAULT_NUM_CORES = 6;
    protected static final int DEFAULT_NUM_PARTITIONS = 3*DEFAULT_NUM_CORES; 
    
    protected String _input_dir;
    protected String _json_list_filename;
    
    protected int    _verbosity;

    public ProcessForWhitelist(String input_dir, String json_list_filename, int verbosity)
    {
        _input_dir = input_dir;
        _json_list_filename = (json_list_filename != null) ? json_list_filename : input_dir;

        _verbosity  = verbosity;
    }

    protected String generateSparkAppName(String exec_mode)
    {
        String spark_app_name = "[" + exec_mode + "] Extracted Features: Process for Whitelist";
        spark_app_name += " [" + _json_list_filename + "]";
    
        return spark_app_name;
    }
        
        
    
    public void execWordCount()
    {   
        String spark_app_name = generateSparkAppName("Per Page");       
        
        SparkConf conf = new SparkConf().setAppName(spark_app_name);
        JavaSparkContext jsc = new JavaSparkContext(conf);

        /*
        if (_verbosity >= 2) {
            System.out.println("Default Minimum Partions: " + jsc.defaultMinPartitions());
            System.out.println("Default Parallelism: " + jsc.defaultParallelism());
        }
            */
        
        int num_partitions = Integer.getInteger("wcsa-ef-ingest.num-partitions", DEFAULT_NUM_PARTITIONS);
        JavaRDD<String> json_list_data = jsc.textFile(_json_list_filename,num_partitions).cache();

        long num_volumes = json_list_data.count();
        double per_vol = 100.0/(double)num_volumes;
        
        //JavaRDD<String> json_list_data_rp = json_list_data.repartition((int)(num_volumes/100));

        DoubleAccumulator per_vol_progress_accum = jsc.sc().doubleAccumulator("Per Volume Progress Percent");
        
        //String strict_file_io_str = System.getProperty("wcsa-ef-ingest.strict-file-io","true");
        boolean strict_file_io = Boolean.getBoolean("wcsa-ef-ingest.strict-file-io");
                
        PerVolumeWordStreamFlatmap paged_solr_wordfreq_flatmap 
            = new PerVolumeWordStreamFlatmap(_input_dir,_verbosity, 
                                     per_vol_progress_accum,per_vol,
                                     strict_file_io);
        JavaRDD<String> words = json_list_data.flatMap(paged_solr_wordfreq_flatmap); // .cache() *****
        
        
        JavaPairRDD<String, Integer> pairs = words.mapToPair(new PairFunction<String, String, Integer>() {
            public Tuple2<String, Integer> call(String s) { return new Tuple2<String, Integer>(s, 1); }
        });

        JavaPairRDD<String, Integer> counts = pairs.reduceByKey(new Function2<Integer, Integer, Integer>() {
            public Integer call(Integer a, Integer b) { return a + b; }
        });

        //counts.map(lambda (x,y): (y,x));
        
        
        JavaPairRDD<Integer, String> swappedPair = counts.mapToPair(new PairFunction<Tuple2<String, Integer>, Integer, String>() {
               @Override
               public Tuple2<Integer, String> call(Tuple2<String, Integer> item) throws Exception {
                   return item.swap();
               }

            });

        //JavaPairRDD<Integer, String> sorted_swapped_pair = swappedPair.sortByKey(false,num_partitions);
        JavaPairRDD<Integer, String> sorted_swapped_pair = swappedPair.sortByKey(false,1);

        JavaPairRDD<String, Integer> sorted_swaped_back_pair = sorted_swapped_pair.mapToPair(new PairFunction<Tuple2<Integer, String>, String, Integer>() {
               @Override
               public Tuple2<String, Integer> call(Tuple2<Integer, String> item) throws Exception {
                   return item.swap();
               }

            });

        /*

        JavaPairRDD<String, Integer> sorted_counts 
            = counts.map(item -> item.swap()) // interchanges position of entries in each tuple
                .sortByKey(true, 1)         // 1st arg configures ascending sort, 2nd arg configures one task
                .map(item -> item.swap());
        
        */
        
        
        //sorted_counts.saveAsTextFile(_json_list_filename + ".out");
        String filename_root = _json_list_filename.replaceAll(".*/","").replaceAll("\\..*$","");
        String output_directory = "whitelist-" + filename_root + "-out";
        sorted_swaped_back_pair.saveAsTextFile(output_directory);

        
        //System.out.println("");
        //System.out.println("############");
        //System.out.println("# Number of page ids: " + num_page_ids);
        //System.out.println("############");
        
        jsc.close();
    }

    
    public static void print_usage(HelpFormatter formatter, Options options)
    {
        formatter.printHelp("RUN.bash [options] input-dir json-filelist.txt", options);
    }
    
    public static void main(String[] args) {
        Options options = new Options();

        Option verbosity_opt = new Option("v", "verbosity", true, 
                "Set to control the level of debugging output [0=none, 1=some, 2=lots]");
        verbosity_opt.setRequired(false);
        options.addOption(verbosity_opt);
        
        Option properties_opt = new Option("p", "properties", true, 
                "Read in the specified Java properties file");
        properties_opt.setRequired(false);
        options.addOption(properties_opt);
        
        // Need to work with CLI v1.2 as this is the JAR that is bundled with Hadoop/Spark 
        CommandLineParser parser = new GnuParser(); 
        //CommandLineParser parser = new DefaultParser(); // if working with CLI v1.3 and above
        
        HelpFormatter formatter = new HelpFormatter();
        CommandLine cmd = null;

        try {
            cmd = parser.parse(options, args);
        }
        catch (ParseException e) {
            System.err.println(e.getMessage());
            print_usage(formatter,options);
            System.exit(1);
        }

        String verbosity_str = cmd.getOptionValue("verbosity","1");
        int verbosity = Integer.parseInt(verbosity_str);

        String property_filename = cmd.getOptionValue("properties",null);
        
        String[] filtered_args = cmd.getArgs();

        if (filtered_args.length != 2) {
            print_usage(formatter,options);
            System.exit(1);
        }
    
        if (property_filename != null) {
            try {
                FileInputStream fis = new FileInputStream(property_filename);
                BufferedInputStream bis = new BufferedInputStream(fis);
                
                System.getProperties().load(bis);
            } 
            catch (FileNotFoundException e) {
                e.printStackTrace();
                System.err.println("File not found: '" + property_filename + "'. Skipping property file read");
            }
            catch (IOException e) {
                System.err.println("IO Exception for: '" + property_filename + "'. Malformed syntax? Skipping property file read");
            }
        }
                
        String input_dir  = filtered_args[0];
        String json_list_filename = filtered_args[1];
        
        ProcessForWhitelist prep_for_whitelist 
            = new ProcessForWhitelist(input_dir,json_list_filename,verbosity);
        
        //String process_ef_json_mode = System.getProperty("wcsa-ef-ingest.process-ef-json-mode","per-page");
        prep_for_whitelist.execWordCount();
    
    }
}