1 | package org.hathitrust;
|
---|
2 |
|
---|
3 | import java.io.BufferedReader;
|
---|
4 | import java.io.BufferedWriter;
|
---|
5 | import java.io.IOException;
|
---|
6 | import java.util.ArrayList;
|
---|
7 | import java.util.Iterator;
|
---|
8 |
|
---|
9 | import org.apache.commons.compress.compressors.CompressorException;
|
---|
10 | import org.apache.spark.api.java.function.FlatMapFunction;
|
---|
11 | import org.json.JSONArray;
|
---|
12 | import org.json.JSONObject;
|
---|
13 |
|
---|
14 | /*
|
---|
15 | class PagedJSON implements Function<String, Boolean> {
|
---|
16 |
|
---|
17 | private static final long serialVersionUID = 1L;
|
---|
18 |
|
---|
19 | public Boolean call(String s) { return s.contains("a"); }
|
---|
20 | }
|
---|
21 | */
|
---|
22 |
|
---|
23 |
|
---|
24 | class PagedJSON implements FlatMapFunction<String, String>
|
---|
25 | {
|
---|
26 | private static final long serialVersionUID = 1L;
|
---|
27 |
|
---|
28 | protected String _input_dir;
|
---|
29 | protected String _output_dir;
|
---|
30 | protected int _verbosity;
|
---|
31 |
|
---|
32 | public PagedJSON(String input_dir, String output_dir, int verbosity)
|
---|
33 | {
|
---|
34 | _input_dir = input_dir;
|
---|
35 | _output_dir = output_dir;
|
---|
36 | _verbosity = verbosity;
|
---|
37 | }
|
---|
38 |
|
---|
39 | protected JSONObject readJSONFile(String filename)
|
---|
40 | {
|
---|
41 | //Path path = Paths.get(filename);
|
---|
42 |
|
---|
43 | StringBuilder sb = new StringBuilder();
|
---|
44 |
|
---|
45 | try {
|
---|
46 |
|
---|
47 | String str;
|
---|
48 | BufferedReader br = ClusterFileIO.getBufferedReaderForCompressedFile(_input_dir + "/" + filename);
|
---|
49 | while ((str = br.readLine()) != null) {
|
---|
50 | sb.append(str);
|
---|
51 | }
|
---|
52 |
|
---|
53 | br.close();
|
---|
54 | }
|
---|
55 | catch (Exception e) {
|
---|
56 | e.printStackTrace();
|
---|
57 | }
|
---|
58 |
|
---|
59 | JSONObject json_obj = new JSONObject(sb.toString());
|
---|
60 |
|
---|
61 | return json_obj;
|
---|
62 | }
|
---|
63 |
|
---|
64 | public Iterator<String> call(String json_file_in)
|
---|
65 | {
|
---|
66 | JSONObject extracted_feature_record = readJSONFile(json_file_in);
|
---|
67 |
|
---|
68 | // Check output directory for volume exists, and create it if not
|
---|
69 |
|
---|
70 |
|
---|
71 | String id = extracted_feature_record.getString("id");
|
---|
72 |
|
---|
73 | //JSONObject ef_metadata = extracted_feature_record.getJSONObject("metadata");
|
---|
74 | JSONObject ef_features = extracted_feature_record.getJSONObject("features");
|
---|
75 |
|
---|
76 |
|
---|
77 | int ef_page_count = ef_features.getInt("pageCount");
|
---|
78 |
|
---|
79 | if (_verbosity >= 1) {
|
---|
80 | System.out.println("Processing: " + json_file_in);
|
---|
81 | System.out.println(" pageCount = " + ef_page_count);
|
---|
82 | }
|
---|
83 |
|
---|
84 | JSONArray ef_pages = ef_features.getJSONArray("pages");
|
---|
85 | int ef_num_pages = ef_pages.length();
|
---|
86 |
|
---|
87 | // Make directory for page-level JSON output
|
---|
88 | String json_dir = ClusterFileIO.removeSuffix(json_file_in,".json.bz2");
|
---|
89 | String page_json_dir = json_dir + "/pages";
|
---|
90 | ClusterFileIO.createDirectoryAll(_output_dir + "/" + page_json_dir);
|
---|
91 | //System.out.println("mkdir: " + _output_dir + "/" + page_json_dir);
|
---|
92 |
|
---|
93 | ArrayList<String> ids = new ArrayList<String>(ef_num_pages);
|
---|
94 | for (int i = 0; i < ef_page_count; i++) {
|
---|
95 | String formatted_i = String.format("page-%06d", i);
|
---|
96 | String page_id = id + "." + formatted_i;
|
---|
97 |
|
---|
98 | if (_verbosity >= 2) {
|
---|
99 | System.out.println(" Page: " + page_id);
|
---|
100 | }
|
---|
101 |
|
---|
102 |
|
---|
103 |
|
---|
104 | String output_json_bz2 = page_json_dir +"/" + formatted_i + ".json.bz2";
|
---|
105 | ids.add(output_json_bz2);
|
---|
106 |
|
---|
107 | if (i==0) {
|
---|
108 | System.out.println("Sample output JSON page file: " + output_json_bz2);
|
---|
109 | }
|
---|
110 |
|
---|
111 | // create JSON obj of just the page (for now), and write it out
|
---|
112 | JSONObject ef_page = ef_pages.getJSONObject(i);
|
---|
113 | try {
|
---|
114 | BufferedWriter bw = ClusterFileIO.getBufferedWriterForCompressedFile(output_json_bz2);
|
---|
115 | bw.write(ef_page.toString());
|
---|
116 | bw.close();
|
---|
117 | } catch (IOException e) {
|
---|
118 | e.printStackTrace();
|
---|
119 | } catch (CompressorException e) {
|
---|
120 | e.printStackTrace();
|
---|
121 | }
|
---|
122 |
|
---|
123 |
|
---|
124 | }
|
---|
125 |
|
---|
126 | /*
|
---|
127 | for (int i = 0; i < ef_num_pages; i++)
|
---|
128 | {
|
---|
129 | //String post_id = ef_pages.getJSONObject(i).getString("post_id");
|
---|
130 | //......
|
---|
131 | }
|
---|
132 | */
|
---|
133 | //String pageName = json_obj.getJSONObject("pageInfo").getString("pageName");
|
---|
134 | /*
|
---|
135 | JSONArray arr = obj.getJSONArray("posts");
|
---|
136 | for (int i = 0; i < arr.length(); i++)
|
---|
137 | {
|
---|
138 | String post_id = arr.getJSONObject(i).getString("post_id");
|
---|
139 | ......
|
---|
140 | }
|
---|
141 | */
|
---|
142 |
|
---|
143 |
|
---|
144 | ids.add(id);
|
---|
145 |
|
---|
146 | return ids.iterator();
|
---|
147 | }
|
---|
148 | }
|
---|
149 |
|
---|