source: other-projects/hathitrust/solr-extracted-features/trunk/src/main/java/org/hathitrust/extractedfeatures/PagedJSON.java@ 30997

Last change on this file since 30997 was 30997, checked in by davidb, 7 years ago

Verbosity control over printing

  • Property svn:executable set to *
File size: 6.3 KB
Line 
1package org.hathitrust.extractedfeatures;
2
3import java.io.BufferedReader;
4import java.io.BufferedWriter;
5import java.io.IOException;
6import java.io.InputStreamReader;
7import java.io.OutputStream;
8import java.net.HttpURLConnection;
9import java.net.URL;
10import java.util.ArrayList;
11import java.util.Set;
12
13import org.apache.commons.compress.compressors.CompressorException;
14import org.apache.spark.api.java.function.FlatMapFunction;
15import org.apache.spark.api.java.function.VoidFunction;
16import org.apache.spark.util.DoubleAccumulator;
17import org.json.JSONArray;
18import org.json.JSONObject;
19
20/*
21class PagedJSON implements Function<String, Boolean> {
22
23 private static final long serialVersionUID = 1L;
24
25 public Boolean call(String s) { return s.contains("a"); }
26}
27 */
28
29
30//class PagedJSON implements FlatMapFunction<String, String>
31public class PagedJSON implements VoidFunction<String>
32{
33 private static final long serialVersionUID = 1L;
34
35 protected String _input_dir;
36 protected String _solr_url;
37 protected String _output_dir;
38 protected int _verbosity;
39
40 DoubleAccumulator _progress_accum;
41 double _progress_step;
42
43 public PagedJSON(String input_dir, String solr_url, String output_dir, int verbosity,
44 DoubleAccumulator progress_accum, double progress_step)
45 {
46 _input_dir = input_dir;
47 _solr_url = solr_url;
48 _output_dir = output_dir;
49 _verbosity = verbosity;
50
51 _progress_accum = progress_accum;
52 _progress_step = progress_step;
53 }
54
55 public static void saveSolrDoc(JSONObject solr_add_doc_json, String output_file_json_bz2)
56 {
57 try {
58 BufferedWriter bw = ClusterFileIO.getBufferedWriterForCompressedFile(output_file_json_bz2);
59 bw.write(solr_add_doc_json.toString());
60 bw.close();
61 } catch (IOException e) {
62 e.printStackTrace();
63 } catch (CompressorException e) {
64 e.printStackTrace();
65 }
66 }
67
68 public static void postSolrDoc(String post_url, JSONObject solr_add_doc_json)
69 {
70
71 //String curl_popen = "curl -X POST -H 'Content-Type: application/json'";
72 //curl_popen += " 'http://10.11.0.53:8983/solr/htrc-pd-ef/update'";
73 //curl_popen += " --data-binary '";
74 //curl_popen += "'"
75
76
77 try {
78 HttpURLConnection httpcon = (HttpURLConnection) ((new URL(post_url).openConnection()));
79 httpcon.setDoOutput(true);
80 httpcon.setRequestProperty("Content-Type", "application/json");
81 httpcon.setRequestProperty("Accept", "application/json");
82 httpcon.setRequestMethod("POST");
83 httpcon.connect();
84
85 byte[] outputBytes = solr_add_doc_json.toString().getBytes("UTF-8");
86 OutputStream os = httpcon.getOutputStream();
87 os.write(outputBytes);
88 os.close();
89
90
91 // Read response
92 StringBuilder sb = new StringBuilder();
93 BufferedReader in = new BufferedReader(new InputStreamReader(httpcon.getInputStream()));
94 String decodedString;
95 while ((decodedString = in.readLine()) != null) {
96 sb.append(decodedString);
97 }
98 in.close();
99
100 JSONObject solr_status_json = new JSONObject(sb.toString());
101 JSONObject response_header_json = solr_status_json.getJSONObject("responseHeader");
102 if (response_header_json != null) {
103 int status = response_header_json.getInt("status");
104 if (status != 0) {
105 System.err.println("Warning: POST request to " + post_url + " returned status " + status);
106 System.err.println("Full response was: " + sb);
107 }
108 }
109 else {
110 System.err.println("Failed response to Solr POST: " + sb);
111 }
112
113
114
115 }
116 catch (Exception e) {
117 e.printStackTrace();
118 }
119
120 }
121
122 //public Iterator<String> call(String json_file_in)
123 public void call(String json_file_in)
124 {
125 JSONObject extracted_feature_record = JSONClusterFileIO.readJSONFile(_input_dir + "/" + json_file_in);
126
127 String volume_id = extracted_feature_record.getString("id");
128
129 //JSONObject ef_metadata = extracted_feature_record.getJSONObject("metadata");
130 //String title= ef_metadata.getString("title");
131
132 JSONObject ef_features = extracted_feature_record.getJSONObject("features");
133
134
135 int ef_page_count = ef_features.getInt("pageCount");
136
137 if (_verbosity >= 1) {
138 System.out.println("Processing: " + json_file_in);
139 System.out.println(" pageCount = " + ef_page_count);
140 }
141
142 JSONArray ef_pages = ef_features.getJSONArray("pages");
143 int ef_num_pages = ef_pages.length();
144
145 // Make directory for page-level JSON output
146 String json_dir = ClusterFileIO.removeSuffix(json_file_in,".json.bz2");
147 String page_json_dir = json_dir + "/pages";
148 ClusterFileIO.createDirectoryAll(_output_dir + "/" + page_json_dir);
149
150 ArrayList<String> ids = new ArrayList<String>(ef_num_pages);
151 for (int i = 0; i < ef_page_count; i++) {
152 String formatted_i = String.format("page-%06d", i);
153 String page_id = volume_id + "." + formatted_i;
154
155 if (_verbosity >= 2) {
156 System.out.println(" Page: " + page_id);
157 }
158
159 String output_json_bz2 = page_json_dir +"/" + formatted_i + ".json.bz2";
160 ids.add(output_json_bz2);
161
162 if (i==0) {
163 System.out.println("Sample output JSON page file: " + output_json_bz2);
164 }
165
166 JSONObject ef_page = ef_pages.getJSONObject(i);
167
168 if (ef_page != null) {
169 // Convert to Solr add form
170 JSONObject solr_add_doc_json = JSONSolrTransform.generateSolrDocJSON(volume_id, page_id, ef_page);
171
172
173 if ((_verbosity >=2) && (i==20)) {
174 System.out.println("==================");
175 System.out.println("Sample output Solr add JSON [page 20]: " + solr_add_doc_json.toString());
176 System.out.println("==================");
177 }
178
179
180 if (_solr_url != null) {
181 if ((_verbosity >=2) && (i==20)) {
182 System.out.println("==================");
183 System.out.println("Posting to: " + _solr_url);
184 System.out.println("==================");
185 }
186 postSolrDoc(_solr_url, solr_add_doc_json);
187 }
188
189 if (_output_dir != null) {
190 if ((_verbosity >=2) && (i==20)) {
191 System.out.println("==================");
192 System.out.println("Saving to: " + _output_dir);
193 System.out.println("==================");
194 }
195 saveSolrDoc(solr_add_doc_json, _output_dir + "/" + output_json_bz2);
196 }
197 }
198 else {
199 System.err.println("Skipping: " + page_id);
200 }
201
202 }
203
204
205 ids.add(volume_id);
206 _progress_accum.add(_progress_step);
207
208 //return ids.iterator();
209 }
210}
211
Note: See TracBrowser for help on using the repository browser.