| 1 | package org.hathitrust.extractedfeatures;
|
|---|
| 2 |
|
|---|
| 3 | import java.util.ArrayList;
|
|---|
| 4 | import java.util.Iterator;
|
|---|
| 5 | import org.apache.spark.api.java.function.FlatMapFunction;
|
|---|
| 6 | import org.apache.spark.api.java.function.VoidFunction;
|
|---|
| 7 | import org.apache.spark.util.DoubleAccumulator;
|
|---|
| 8 | import org.json.JSONArray;
|
|---|
| 9 | import org.json.JSONObject;
|
|---|
| 10 |
|
|---|
| 11 | /*
|
|---|
| 12 | class PagedJSON implements Function<String, Boolean> {
|
|---|
| 13 |
|
|---|
| 14 | private static final long serialVersionUID = 1L;
|
|---|
| 15 |
|
|---|
| 16 | public Boolean call(String s) { return s.contains("a"); }
|
|---|
| 17 | }
|
|---|
| 18 | */
|
|---|
| 19 |
|
|---|
| 20 |
|
|---|
| 21 | class PagedJSONForeach extends BasePerJSON implements VoidFunction<JSONObject>
|
|---|
| 22 | //public class PagedJSON implements VoidFunction<String>
|
|---|
| 23 | {
|
|---|
| 24 | private static final long serialVersionUID = 1L;
|
|---|
| 25 |
|
|---|
| 26 | public PagedJSONForeach()
|
|---|
| 27 | {
|
|---|
| 28 | super();
|
|---|
| 29 | }
|
|---|
| 30 |
|
|---|
| 31 | public PagedJSONForeach(String input_dir, String solr_url, String output_dir, int verbosity,
|
|---|
| 32 | DoubleAccumulator progress_accum, double progress_step)
|
|---|
| 33 | {
|
|---|
| 34 | super(input_dir,solr_url,output_dir,verbosity,progress_accum,progress_step);
|
|---|
| 35 | }
|
|---|
| 36 |
|
|---|
| 37 |
|
|---|
| 38 | public void call(JSONObject solr_add_doc_json)
|
|---|
| 39 | {
|
|---|
| 40 | String output_json_bz2 = solr_add_doc_json.getString("filename_json_bz2");
|
|---|
| 41 | solr_add_doc_json.remove("filename_json_bz2");
|
|---|
| 42 |
|
|---|
| 43 | boolean random_test = (Math.random()>0.999); // every 1000
|
|---|
| 44 |
|
|---|
| 45 | if ((_verbosity >=2) && (random_test)) {
|
|---|
| 46 | System.out.println("==================");
|
|---|
| 47 | System.out.println("Sample output Solr add JSON [random test 1/1000]: " + solr_add_doc_json.toString());
|
|---|
| 48 | System.out.println("==================");
|
|---|
| 49 | }
|
|---|
| 50 |
|
|---|
| 51 |
|
|---|
| 52 | if (_solr_url != null) {
|
|---|
| 53 | if ((_verbosity >=2) && (random_test)) {
|
|---|
| 54 | System.out.println("==================");
|
|---|
| 55 | System.out.println("Posting to: " + _solr_url);
|
|---|
| 56 | System.out.println("==================");
|
|---|
| 57 | }
|
|---|
| 58 | JSONSolrTransform.postSolrDoc(_solr_url, solr_add_doc_json);
|
|---|
| 59 | }
|
|---|
| 60 |
|
|---|
| 61 | if (_output_dir != null) {
|
|---|
| 62 | if ((_verbosity >=2) && (random_test)) {
|
|---|
| 63 | System.out.println("==================");
|
|---|
| 64 | System.out.println("Saving to: " + _output_dir);
|
|---|
| 65 | System.out.println("==================");
|
|---|
| 66 | }
|
|---|
| 67 | JSONSolrTransform.saveSolrDoc(solr_add_doc_json, _output_dir + "/" + output_json_bz2);
|
|---|
| 68 | }
|
|---|
| 69 | }
|
|---|
| 70 |
|
|---|
| 71 | }
|
|---|
| 72 |
|
|---|
