[31007] | 1 | package org.hathitrust.extractedfeatures;
|
---|
| 2 |
|
---|
| 3 | import java.io.BufferedReader;
|
---|
| 4 | import java.io.BufferedWriter;
|
---|
| 5 | import java.io.IOException;
|
---|
| 6 | import java.io.InputStreamReader;
|
---|
| 7 | import java.io.OutputStream;
|
---|
[31243] | 8 | import java.io.Reader;
|
---|
| 9 | import java.io.StringReader;
|
---|
[31007] | 10 | import java.net.HttpURLConnection;
|
---|
| 11 | import java.net.URL;
|
---|
[31176] | 12 | import java.util.ArrayList;
|
---|
[31007] | 13 | import java.util.Iterator;
|
---|
[31243] | 14 | import java.util.Set;
|
---|
[31007] | 15 |
|
---|
| 16 | import org.apache.commons.compress.compressors.CompressorException;
|
---|
| 17 | import org.json.JSONObject;
|
---|
[31244] | 18 | import org.apache.lucene.analysis.TokenStream;
|
---|
[31243] | 19 | import org.apache.lucene.analysis.Tokenizer;
|
---|
| 20 | import org.apache.lucene.analysis.icu.segmentation.ICUTokenizer;
|
---|
| 21 | import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
|
---|
| 22 | import org.apache.lucene.analysis.tokenattributes.OffsetAttribute;
|
---|
[31007] | 23 |
|
---|
| 24 | public class SolrDocJSON {
|
---|
| 25 |
|
---|
[31245] | 26 | protected static ArrayList<String> getTokenPosCountWords(JSONObject ef_token_pos_count, String page_id)
|
---|
| 27 | {
|
---|
| 28 | boolean solr_icu_tokenize = true;
|
---|
[31243] | 29 |
|
---|
[31245] | 30 | ArrayList<String> words = new ArrayList<String>();
|
---|
| 31 |
|
---|
| 32 | if (ef_token_pos_count != null) {
|
---|
| 33 |
|
---|
[31007] | 34 | Iterator<String> token_iter = ef_token_pos_count.keys();
|
---|
[31243] | 35 | while (token_iter.hasNext()) {
|
---|
| 36 | String token = token_iter.next();
|
---|
[31245] | 37 |
|
---|
[31243] | 38 | if (solr_icu_tokenize == true) {
|
---|
| 39 | Reader reader = new StringReader(token);
|
---|
| 40 |
|
---|
| 41 | Tokenizer tokenizer = new ICUTokenizer();
|
---|
| 42 | tokenizer.setReader(reader);
|
---|
| 43 |
|
---|
| 44 | CharTermAttribute charTermAttribute = tokenizer.addAttribute(CharTermAttribute.class);
|
---|
| 45 |
|
---|
| 46 | try {
|
---|
| 47 | tokenizer.reset();
|
---|
| 48 |
|
---|
| 49 | while (tokenizer.incrementToken()) {
|
---|
| 50 | String term = charTermAttribute.toString();
|
---|
[31245] | 51 | words.add(term);
|
---|
[31243] | 52 | }
|
---|
| 53 |
|
---|
[31244] | 54 | tokenizer.end();
|
---|
[31243] | 55 | tokenizer.close();
|
---|
| 56 | }
|
---|
| 57 | catch (IOException e) {
|
---|
| 58 | e.printStackTrace();
|
---|
| 59 | }
|
---|
| 60 | }
|
---|
| 61 | else {
|
---|
[31245] | 62 | words.add(token);
|
---|
[31243] | 63 | }
|
---|
[31245] | 64 | }
|
---|
| 65 | }
|
---|
| 66 | else {
|
---|
| 67 | System.err.println("Warning: empty tokenPosCount field for '" + page_id + "'");
|
---|
| 68 | }
|
---|
| 69 |
|
---|
| 70 | /* Alternative way to get at keys
|
---|
| 71 | Set<String> token_keys = ef_token_pos_count.keySet();
|
---|
| 72 | for (String token : token_keys) {
|
---|
| 73 | sb.append(token + " ");
|
---|
| 74 | }
|
---|
| 75 | */
|
---|
| 76 | return words;
|
---|
| 77 | }
|
---|
| 78 |
|
---|
| 79 |
|
---|
| 80 | protected static String generateSolrText(JSONObject ef_token_pos_count, String page_id,
|
---|
| 81 | WhitelistBloomFilter whitelist_bloomfilter)
|
---|
| 82 | {
|
---|
| 83 | ArrayList<String> tokens = getTokenPosCountWords(ef_token_pos_count, page_id);
|
---|
| 84 |
|
---|
| 85 | StringBuilder sb = new StringBuilder();
|
---|
| 86 |
|
---|
| 87 | if (whitelist_bloomfilter == null) {
|
---|
| 88 |
|
---|
| 89 | boolean first_append = true;
|
---|
| 90 |
|
---|
| 91 | for (int i=0; i<tokens.size(); i++) {
|
---|
| 92 | String token = tokens.get(i);
|
---|
| 93 |
|
---|
| 94 | if (!first_append) {
|
---|
| 95 | sb.append(" ");
|
---|
| 96 | }
|
---|
| 97 | else {
|
---|
| 98 | first_append = false;
|
---|
| 99 | }
|
---|
| 100 | sb.append(token);
|
---|
[31243] | 101 | }
|
---|
[31245] | 102 | }
|
---|
| 103 | else {
|
---|
| 104 | boolean first_append = true;
|
---|
[31220] | 105 |
|
---|
[31245] | 106 | for (int i=0; i<tokens.size(); i++) {
|
---|
| 107 | String token = tokens.get(i);
|
---|
| 108 |
|
---|
| 109 | if (whitelist_bloomfilter.contains(token)) {
|
---|
[31243] | 110 | if (!first_append) {
|
---|
[31220] | 111 | sb.append(" ");
|
---|
| 112 | }
|
---|
[31243] | 113 | else {
|
---|
| 114 | first_append = false;
|
---|
| 115 | }
|
---|
| 116 | sb.append(token);
|
---|
[31245] | 117 | }
|
---|
[31007] | 118 | }
|
---|
[31245] | 119 |
|
---|
[31007] | 120 | }
|
---|
| 121 |
|
---|
[31245] | 122 |
|
---|
| 123 | return sb.toString();
|
---|
| 124 | }
|
---|
| 125 |
|
---|
[31220] | 126 | protected static JSONObject generateSolrDocJSON(String volume_id, String page_id, JSONObject ef_page,
|
---|
| 127 | WhitelistBloomFilter whitelist_bloomfilter)
|
---|
[31007] | 128 | {
|
---|
| 129 | JSONObject solr_update_json = null;
|
---|
| 130 |
|
---|
| 131 | if (ef_page != null) {
|
---|
| 132 | JSONObject ef_body = ef_page.getJSONObject("body");
|
---|
| 133 | if (ef_body != null) {
|
---|
| 134 | JSONObject ef_token_pos_count = ef_body.getJSONObject("tokenPosCount");
|
---|
| 135 | if (ef_token_pos_count != null) {
|
---|
| 136 |
|
---|
| 137 | JSONObject solr_add_json = new JSONObject();
|
---|
| 138 |
|
---|
[31245] | 139 | String text = generateSolrText(ef_token_pos_count,page_id,whitelist_bloomfilter);
|
---|
[31007] | 140 |
|
---|
| 141 | JSONObject solr_doc_json = new JSONObject();
|
---|
| 142 | solr_doc_json.put("id", page_id);
|
---|
| 143 | solr_doc_json.put("volumeid_s", volume_id);
|
---|
[31220] | 144 | if (!text.equals("")) {
|
---|
| 145 | solr_doc_json.put("eftext_txt", text);
|
---|
| 146 | }
|
---|
| 147 | else {
|
---|
| 148 | solr_doc_json.put("efnotext_b", true);
|
---|
| 149 | }
|
---|
[31007] | 150 | solr_add_json.put("commitWithin", 5000);
|
---|
| 151 | solr_add_json.put("doc", solr_doc_json);
|
---|
| 152 |
|
---|
| 153 | solr_update_json = new JSONObject();
|
---|
| 154 | solr_update_json.put("add",solr_add_json);
|
---|
| 155 |
|
---|
| 156 | }
|
---|
| 157 | else {
|
---|
| 158 | System.err.println("Warning: empty tokenPosCount field for '" + page_id + "'");
|
---|
| 159 | }
|
---|
| 160 | }
|
---|
| 161 | else {
|
---|
| 162 | System.err.println("Warning: empty body field for '" + page_id + "'");
|
---|
| 163 | }
|
---|
| 164 |
|
---|
| 165 | }
|
---|
| 166 | else {
|
---|
| 167 | System.err.println("Warning: null page for '" + page_id + "'");
|
---|
| 168 | }
|
---|
| 169 |
|
---|
| 170 |
|
---|
| 171 | /*
|
---|
| 172 |
|
---|
| 173 | /update/json/docs
|
---|
| 174 | */
|
---|
| 175 |
|
---|
| 176 | // For Reference ...
|
---|
| 177 | // Example documentation on Solr JSON syntax:
|
---|
| 178 | // https://cwiki.apache.org/confluence/display/solr/Uploading+Data+with+Index+Handlers
|
---|
| 179 | // #UploadingDatawithIndexHandlers-JSONFormattedIndexUpdates
|
---|
| 180 |
|
---|
| 181 | /*
|
---|
| 182 | curl -X POST -H 'Content-Type: application/json' 'http://localhost:8983/solr/my_collection/update' --data-binary '
|
---|
| 183 | {
|
---|
| 184 | "add": {
|
---|
| 185 | "doc": {
|
---|
| 186 | "id": "DOC1",
|
---|
| 187 | "my_boosted_field": { use a map with boost/value for a boosted field
|
---|
| 188 | "boost": 2.3,
|
---|
| 189 | "value": "test"
|
---|
| 190 | },
|
---|
| 191 | "my_multivalued_field": [ "aaa", "bbb" ] Can use an array for a multi-valued field
|
---|
| 192 | }
|
---|
| 193 | },
|
---|
| 194 | "add": {
|
---|
| 195 | "commitWithin": 5000, commit this document within 5 seconds
|
---|
| 196 | "overwrite": false, don't check for existing documents with the same uniqueKey
|
---|
| 197 | "boost": 3.45, a document boost
|
---|
| 198 | "doc": {
|
---|
| 199 | "f1": "v1", Can use repeated keys for a multi-valued field
|
---|
| 200 | "f1": "v2"
|
---|
| 201 | }
|
---|
| 202 | },
|
---|
| 203 |
|
---|
| 204 | "commit": {},
|
---|
| 205 | "optimize": { "waitSearcher":false },
|
---|
| 206 |
|
---|
| 207 | "delete": { "id":"ID" }, delete by ID
|
---|
| 208 | "delete": { "query":"QUERY" } delete by query
|
---|
| 209 | }'
|
---|
| 210 | */
|
---|
| 211 |
|
---|
| 212 | return solr_update_json;
|
---|
| 213 | }
|
---|
| 214 |
|
---|
[31242] | 215 | protected static ArrayList<String> generateTokenPosCountText(String volume_id, String page_id, JSONObject ef_page)
|
---|
[31176] | 216 | {
|
---|
[31245] | 217 | ArrayList<String> word_list = null;
|
---|
[31176] | 218 |
|
---|
| 219 | if (ef_page != null) {
|
---|
| 220 | JSONObject ef_body = ef_page.getJSONObject("body");
|
---|
| 221 | if (ef_body != null) {
|
---|
| 222 | JSONObject ef_token_pos_count = ef_body.getJSONObject("tokenPosCount");
|
---|
[31245] | 223 | word_list = getTokenPosCountWords(ef_token_pos_count,page_id);
|
---|
[31176] | 224 | }
|
---|
| 225 | else {
|
---|
| 226 | System.err.println("Warning: empty body field for '" + page_id + "'");
|
---|
| 227 | }
|
---|
| 228 |
|
---|
| 229 | }
|
---|
| 230 | else {
|
---|
| 231 | System.err.println("Warning: null page for '" + page_id + "'");
|
---|
| 232 | }
|
---|
| 233 |
|
---|
| 234 | return word_list;
|
---|
| 235 | }
|
---|
| 236 |
|
---|
[31007] | 237 | public static void saveSolrDoc(JSONObject solr_add_doc_json, String output_file_json_bz2)
|
---|
| 238 | {
|
---|
| 239 | try {
|
---|
| 240 | BufferedWriter bw = ClusterFileIO.getBufferedWriterForCompressedFile(output_file_json_bz2);
|
---|
| 241 | bw.write(solr_add_doc_json.toString());
|
---|
| 242 | bw.close();
|
---|
| 243 | } catch (IOException e) {
|
---|
| 244 | e.printStackTrace();
|
---|
| 245 | } catch (CompressorException e) {
|
---|
| 246 | e.printStackTrace();
|
---|
| 247 | }
|
---|
| 248 | }
|
---|
| 249 |
|
---|
| 250 | public static void postSolrDoc(String post_url, JSONObject solr_add_doc_json)
|
---|
| 251 | {
|
---|
| 252 |
|
---|
| 253 | //String curl_popen = "curl -X POST -H 'Content-Type: application/json'";
|
---|
| 254 | //curl_popen += " 'http://10.11.0.53:8983/solr/htrc-pd-ef/update'";
|
---|
| 255 | //curl_popen += " --data-binary '";
|
---|
| 256 | //curl_popen += "'"
|
---|
| 257 |
|
---|
| 258 |
|
---|
| 259 | try {
|
---|
| 260 | HttpURLConnection httpcon = (HttpURLConnection) ((new URL(post_url).openConnection()));
|
---|
| 261 | httpcon.setDoOutput(true);
|
---|
| 262 | httpcon.setRequestProperty("Content-Type", "application/json");
|
---|
| 263 | httpcon.setRequestProperty("Accept", "application/json");
|
---|
| 264 | httpcon.setRequestMethod("POST");
|
---|
| 265 | httpcon.connect();
|
---|
| 266 |
|
---|
| 267 | byte[] outputBytes = solr_add_doc_json.toString().getBytes("UTF-8");
|
---|
| 268 | OutputStream os = httpcon.getOutputStream();
|
---|
| 269 | os.write(outputBytes);
|
---|
| 270 | os.close();
|
---|
| 271 |
|
---|
| 272 |
|
---|
| 273 | // Read response
|
---|
| 274 | StringBuilder sb = new StringBuilder();
|
---|
| 275 | BufferedReader in = new BufferedReader(new InputStreamReader(httpcon.getInputStream()));
|
---|
| 276 | String decodedString;
|
---|
| 277 | while ((decodedString = in.readLine()) != null) {
|
---|
| 278 | sb.append(decodedString);
|
---|
| 279 | }
|
---|
| 280 | in.close();
|
---|
| 281 |
|
---|
| 282 | JSONObject solr_status_json = new JSONObject(sb.toString());
|
---|
| 283 | JSONObject response_header_json = solr_status_json.getJSONObject("responseHeader");
|
---|
| 284 | if (response_header_json != null) {
|
---|
| 285 | int status = response_header_json.getInt("status");
|
---|
| 286 | if (status != 0) {
|
---|
| 287 | System.err.println("Warning: POST request to " + post_url + " returned status " + status);
|
---|
| 288 | System.err.println("Full response was: " + sb);
|
---|
| 289 | }
|
---|
| 290 | }
|
---|
| 291 | else {
|
---|
| 292 | System.err.println("Failed response to Solr POST: " + sb);
|
---|
| 293 | }
|
---|
| 294 |
|
---|
| 295 |
|
---|
| 296 |
|
---|
| 297 | }
|
---|
| 298 | catch (Exception e) {
|
---|
| 299 | e.printStackTrace();
|
---|
| 300 | }
|
---|
| 301 |
|
---|
| 302 | }
|
---|
| 303 | }
|
---|