[31001] | 1 | package org.hathitrust.extractedfeatures;
|
---|
| 2 |
|
---|
| 3 | import org.apache.spark.api.java.function.VoidFunction;
|
---|
| 4 | import org.apache.spark.util.DoubleAccumulator;
|
---|
| 5 | import org.json.JSONArray;
|
---|
| 6 | import org.json.JSONObject;
|
---|
| 7 |
|
---|
| 8 | /*
|
---|
| 9 | class PagedJSON implements Function<String, Boolean> {
|
---|
| 10 |
|
---|
| 11 | private static final long serialVersionUID = 1L;
|
---|
| 12 |
|
---|
| 13 | public Boolean call(String s) { return s.contains("a"); }
|
---|
| 14 | }
|
---|
| 15 | */
|
---|
| 16 |
|
---|
| 17 |
|
---|
[31005] | 18 | public class PerVolumeJSON implements VoidFunction<String>
|
---|
[31001] | 19 | {
|
---|
| 20 | private static final long serialVersionUID = 1L;
|
---|
[31005] | 21 | protected String _input_dir;
|
---|
| 22 | protected String _solr_url;
|
---|
| 23 | protected String _output_dir;
|
---|
| 24 | protected int _verbosity;
|
---|
[31001] | 25 |
|
---|
[31005] | 26 | protected DoubleAccumulator _progress_accum;
|
---|
| 27 | protected double _progress_step;
|
---|
[31003] | 28 |
|
---|
[31001] | 29 | public PerVolumeJSON(String input_dir, String solr_url, String output_dir, int verbosity,
|
---|
[31005] | 30 | DoubleAccumulator progress_accum, double progress_step)
|
---|
[31001] | 31 | {
|
---|
[31005] | 32 | _input_dir = input_dir;
|
---|
| 33 | _solr_url = solr_url;
|
---|
| 34 | _output_dir = output_dir;
|
---|
| 35 | _verbosity = verbosity;
|
---|
| 36 |
|
---|
| 37 | _progress_accum = progress_accum;
|
---|
| 38 | _progress_step = progress_step;
|
---|
[31001] | 39 | }
|
---|
[31005] | 40 |
|
---|
[31001] | 41 | //public Iterator<String> call(String json_file_in)
|
---|
| 42 | public void call(String json_file_in)
|
---|
| 43 | {
|
---|
| 44 | JSONObject extracted_feature_record = JSONClusterFileIO.readJSONFile(_input_dir + "/" + json_file_in);
|
---|
| 45 |
|
---|
| 46 | String volume_id = extracted_feature_record.getString("id");
|
---|
| 47 |
|
---|
| 48 | //JSONObject ef_metadata = extracted_feature_record.getJSONObject("metadata");
|
---|
| 49 | //String title= ef_metadata.getString("title");
|
---|
| 50 |
|
---|
| 51 | JSONObject ef_features = extracted_feature_record.getJSONObject("features");
|
---|
| 52 |
|
---|
| 53 |
|
---|
| 54 | int ef_page_count = ef_features.getInt("pageCount");
|
---|
| 55 |
|
---|
| 56 | if (_verbosity >= 1) {
|
---|
| 57 | System.out.println("Processing: " + json_file_in);
|
---|
| 58 | System.out.println(" pageCount = " + ef_page_count);
|
---|
| 59 | }
|
---|
| 60 |
|
---|
| 61 | JSONArray ef_pages = ef_features.getJSONArray("pages");
|
---|
| 62 | int ef_num_pages = ef_pages.length();
|
---|
| 63 |
|
---|
| 64 | // Make directory for page-level JSON output
|
---|
| 65 | String json_dir = ClusterFileIO.removeSuffix(json_file_in,".json.bz2");
|
---|
| 66 | String page_json_dir = json_dir + "/pages";
|
---|
| 67 |
|
---|
[31096] | 68 | if (_output_dir != null) {
|
---|
| 69 | ClusterFileIO.createDirectoryAll(_output_dir + "/" + page_json_dir);
|
---|
| 70 | }
|
---|
| 71 |
|
---|
[31001] | 72 | //ArrayList<String> ids = new ArrayList<String>(ef_num_pages);
|
---|
| 73 | for (int i = 0; i < ef_page_count; i++) {
|
---|
| 74 | String formatted_i = String.format("page-%06d", i);
|
---|
| 75 | String page_id = volume_id + "." + formatted_i;
|
---|
| 76 |
|
---|
| 77 | if (_verbosity >= 2) {
|
---|
| 78 | System.out.println(" Page: " + page_id);
|
---|
| 79 | }
|
---|
| 80 |
|
---|
| 81 | String output_json_bz2 = page_json_dir +"/" + formatted_i + ".json.bz2";
|
---|
| 82 | //ids.add(output_json_bz2); // ****
|
---|
| 83 |
|
---|
| 84 | if (i==0) {
|
---|
| 85 | System.out.println("Sample output JSON page file: " + output_json_bz2);
|
---|
| 86 | }
|
---|
| 87 |
|
---|
| 88 | JSONObject ef_page = ef_pages.getJSONObject(i);
|
---|
| 89 |
|
---|
| 90 | if (ef_page != null) {
|
---|
| 91 | // Convert to Solr add form
|
---|
[31007] | 92 | JSONObject solr_add_doc_json = SolrDocJSON.generateSolrDocJSON(volume_id, page_id, ef_page);
|
---|
[31001] | 93 |
|
---|
| 94 |
|
---|
| 95 | if ((_verbosity >=2) && (i==20)) {
|
---|
| 96 | System.out.println("==================");
|
---|
| 97 | System.out.println("Sample output Solr add JSON [page 20]: " + solr_add_doc_json.toString());
|
---|
| 98 | System.out.println("==================");
|
---|
| 99 | }
|
---|
| 100 |
|
---|
| 101 |
|
---|
| 102 | if (_solr_url != null) {
|
---|
| 103 | if ((_verbosity >=2) && (i==20)) {
|
---|
| 104 | System.out.println("==================");
|
---|
| 105 | System.out.println("Posting to: " + _solr_url);
|
---|
| 106 | System.out.println("==================");
|
---|
| 107 | }
|
---|
[31007] | 108 | SolrDocJSON.postSolrDoc(_solr_url, solr_add_doc_json);
|
---|
[31001] | 109 | }
|
---|
| 110 |
|
---|
| 111 | if (_output_dir != null) {
|
---|
| 112 | if ((_verbosity >=2) && (i==20)) {
|
---|
| 113 | System.out.println("==================");
|
---|
| 114 | System.out.println("Saving to: " + _output_dir);
|
---|
| 115 | System.out.println("==================");
|
---|
| 116 | }
|
---|
[31007] | 117 | SolrDocJSON.saveSolrDoc(solr_add_doc_json, _output_dir + "/" + output_json_bz2);
|
---|
[31001] | 118 | }
|
---|
| 119 | }
|
---|
| 120 | else {
|
---|
| 121 | System.err.println("Skipping: " + page_id);
|
---|
| 122 | }
|
---|
| 123 |
|
---|
| 124 | }
|
---|
| 125 |
|
---|
| 126 |
|
---|
| 127 | //ids.add(volume_id);
|
---|
| 128 | _progress_accum.add(_progress_step);
|
---|
| 129 |
|
---|
| 130 | //return ids.iterator();
|
---|
| 131 | }
|
---|
| 132 | }
|
---|
| 133 |
|
---|