package org.hathitrust.extractedfeatures; import java.util.ArrayList; import java.util.Iterator; import org.apache.spark.api.java.function.FlatMapFunction; 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 { private static final long serialVersionUID = 1L; public Boolean call(String s) { return s.contains("a"); } } */ class __PerPageJSONForeach implements VoidFunction //public class PagedJSON implements VoidFunction { private static final long serialVersionUID = 1L; protected String _input_dir; protected String _solr_url; protected String _output_dir; protected int _verbosity; protected DoubleAccumulator _progress_accum; protected double _progress_step; public __PerPageJSONForeach(String input_dir, String solr_url, String output_dir, int verbosity, DoubleAccumulator progress_accum, double progress_step) { _input_dir = input_dir; _solr_url = solr_url; _output_dir = output_dir; _verbosity = verbosity; _progress_accum = progress_accum; _progress_step = progress_step; } public void call(JSONObject solr_add_doc_json) { String output_json_bz2 = solr_add_doc_json.getString("filename_json_bz2"); solr_add_doc_json.remove("filename_json_bz2"); boolean random_test = (Math.random()>0.999); // every 1000 if ((_verbosity >=2) && (random_test)) { System.out.println("=================="); System.out.println("Sample output Solr add JSON [random test 1/1000]: " + solr_add_doc_json.toString()); System.out.println("=================="); } if (_solr_url != null) { if ((_verbosity >=2) && (random_test)) { 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) && (random_test)) { System.out.println("=================="); System.out.println("Saving to: " + _output_dir); System.out.println("=================="); } SolrDocJSON.saveSolrDoc(solr_add_doc_json, _output_dir + "/" + output_json_bz2); } } }