source: other-projects/hathitrust/wcsa/extracted-features-solr/trunk/solr-ingest/src/main/java/org/hathitrust/extractedfeatures/PerVolumeJSON.java@ 31225

package org.hathitrust.extractedfeatures;

import org.apache.spark.api.java.function.VoidFunction;
import org.apache.spark.util.DoubleAccumulator;
import org.json.JSONArray;
import org.json.JSONObject;

/*
class PagedJSON implements Function<String, Boolean> {

    private static final long serialVersionUID = 1L;

    public Boolean call(String s) { return s.contains("a"); }
}
 */


public class PerVolumeJSON implements VoidFunction<String> 
{
    private static final long serialVersionUID = 1L;
    protected String _input_dir;
    protected String _whitelist_filename;
    
    protected String _solr_url;
    protected String _output_dir;
    
    protected int    _verbosity;
    
    protected WhitelistBloomFilter _whitelist_bloomfilter;
    
    protected DoubleAccumulator _progress_accum;
    protected double            _progress_step;
    
    public PerVolumeJSON(String input_dir, String whitelist_filename,
                         String solr_url, String output_dir, int verbosity, 
                         DoubleAccumulator progress_accum, double progress_step)
    {
        _input_dir  = input_dir;
        _whitelist_filename = whitelist_filename;
        
        _solr_url   = solr_url;
        _output_dir = output_dir;
        _verbosity  = verbosity;
        
        _progress_accum = progress_accum;
        _progress_step  = progress_step;
        
        _whitelist_bloomfilter = null;
    }
    
    //public Iterator<String> call(String json_file_in) 
    public void call(String json_file_in) 
    { 
        if ((_whitelist_filename != null) && (_whitelist_bloomfilter != null)) {
            _whitelist_bloomfilter = new WhitelistBloomFilter(_whitelist_filename,true);
        }

        JSONObject extracted_feature_record = JSONClusterFileIO.readJSONFile(_input_dir + "/" + json_file_in);
        
        String volume_id = extracted_feature_record.getString("id");
        
        //JSONObject ef_metadata = extracted_feature_record.getJSONObject("metadata");
        //String title= ef_metadata.getString("title");
        
        JSONObject ef_features = extracted_feature_record.getJSONObject("features");
        
        
        int ef_page_count = ef_features.getInt("pageCount");
        
        if (_verbosity >= 1) {
            System.out.println("Processing: " + json_file_in);
            System.out.println("  pageCount = " + ef_page_count);
        }
        
        JSONArray ef_pages = ef_features.getJSONArray("pages");
        int ef_num_pages = ef_pages.length();
        
        // Make directory for page-level JSON output
        String json_dir = ClusterFileIO.removeSuffix(json_file_in,".json.bz2");
        String page_json_dir = json_dir + "/pages";
        
        if (_output_dir != null) {
            ClusterFileIO.createDirectoryAll(_output_dir + "/" + page_json_dir);
        }
        
        //ArrayList<String> ids = new ArrayList<String>(ef_num_pages);
        for (int i = 0; i < ef_page_count; i++) {
            String formatted_i = String.format("page-%06d", i);
            String page_id = volume_id + "." + formatted_i;
        
            if (_verbosity >= 2) {
              System.out.println("  Page: " + page_id);
            }
            
            String output_json_bz2 = page_json_dir +"/" + formatted_i + ".json.bz2";
            //ids.add(output_json_bz2); // ****
            
            if (i==0) {
                System.out.println("Sample output JSON page file: " + output_json_bz2);
            }
            
            JSONObject ef_page = ef_pages.getJSONObject(i);

            if (ef_page != null) {
                // Convert to Solr add form
                JSONObject solr_add_doc_json 
                    = SolrDocJSON.generateSolrDocJSON(volume_id, page_id, ef_page, _whitelist_bloomfilter);

            
                if ((_verbosity >=2) && (i==20)) {
                    System.out.println("==================");
                    System.out.println("Sample output Solr add JSON [page 20]: " + solr_add_doc_json.toString());
                    System.out.println("==================");
                }
                
                            
                if (_solr_url != null) {
                    if ((_verbosity >=2) && (i==20)) {
                        System.out.println("==================");
                        System.out.println("Posting to: " + _solr_url);
                        System.out.println("==================");
                    }
                    SolrDocJSON.postSolrDoc(_solr_url, solr_add_doc_json);
                }

                if (_output_dir != null) {
                    if ((_verbosity >=2) && (i==20)) {
                        System.out.println("==================");
                        System.out.println("Saving to: " + _output_dir);
                        System.out.println("==================");
                    }
                    SolrDocJSON.saveSolrDoc(solr_add_doc_json, _output_dir + "/" + output_json_bz2);
                }
            }
            else {
                System.err.println("Skipping: " + page_id);
            }
            
        }
        
        
        //ids.add(volume_id);
        _progress_accum.add(_progress_step);
        
        //return ids.iterator();
    }
}