1 | package eu.clarin.sru.fcs.aggregator.search; |
---|
2 | |
---|
3 | import eu.clarin.sru.client.SRUVersion; |
---|
4 | import java.util.List; |
---|
5 | import eu.clarin.sru.client.SRUClientException; |
---|
6 | import eu.clarin.sru.client.SRUSearchRetrieveRequest; |
---|
7 | import eu.clarin.sru.client.SRUSearchRetrieveResponse; |
---|
8 | import eu.clarin.sru.client.fcs.ClarinFCSRecordData; |
---|
9 | import eu.clarin.sru.fcs.aggregator.client.ThrottledClient; |
---|
10 | import eu.clarin.sru.fcs.aggregator.scan.Corpus; |
---|
11 | import eu.clarin.sru.fcs.aggregator.scan.FCSProtocolVersion; |
---|
12 | import eu.clarin.sru.fcs.aggregator.scan.Statistics; |
---|
13 | import eu.clarin.sru.fcs.aggregator.util.SRUCQL; |
---|
14 | import java.util.ArrayList; |
---|
15 | import java.util.Collections; |
---|
16 | import java.util.Random; |
---|
17 | import java.util.concurrent.atomic.AtomicLong; |
---|
18 | import java.util.logging.Level; |
---|
19 | import java.util.logging.Logger; |
---|
20 | import opennlp.tools.tokenize.TokenizerModel; |
---|
21 | import org.slf4j.LoggerFactory; |
---|
22 | |
---|
23 | /** |
---|
24 | * Class representing a search operation |
---|
25 | * |
---|
26 | * @author Yana Panchenko |
---|
27 | * @author edima |
---|
28 | */ |
---|
29 | public class Search { |
---|
30 | |
---|
31 | private static final org.slf4j.Logger log = LoggerFactory.getLogger(Search.class); |
---|
32 | |
---|
33 | private static final String SEARCH_RESULTS_ENCODING = "UTF-8"; |
---|
34 | |
---|
35 | private static final AtomicLong counter = new AtomicLong(Math.abs(new Random().nextInt())); |
---|
36 | |
---|
37 | private final Long id; |
---|
38 | private final long createdAt = System.currentTimeMillis(); |
---|
39 | private final String searchLanguage; |
---|
40 | private final List<Request> requests = Collections.synchronizedList(new ArrayList<Request>()); |
---|
41 | private final List<Result> results = Collections.synchronizedList(new ArrayList<Result>()); |
---|
42 | private final Statistics statistics; |
---|
43 | |
---|
44 | public Search(ThrottledClient searchClient, SRUVersion version, |
---|
45 | Statistics statistics, List<Corpus> corpora, String searchString, |
---|
46 | String searchLanguage, int startRecord, int maxRecords |
---|
47 | ) { |
---|
48 | this.id = counter.getAndIncrement(); |
---|
49 | this.searchLanguage = searchLanguage; |
---|
50 | this.statistics = statistics; |
---|
51 | for (Corpus corpus : corpora) { |
---|
52 | executeSearch(searchClient, version, corpus, searchString, startRecord, maxRecords); |
---|
53 | } |
---|
54 | } |
---|
55 | |
---|
56 | private Request executeSearch(ThrottledClient searchClient, SRUVersion version, final Corpus corpus, String searchString, int startRecord, int maxRecords) { |
---|
57 | final Request request = new Request(corpus, searchString, startRecord, startRecord + maxRecords - 1); |
---|
58 | log.info("Executing search in '{}' query='{}' maxRecords='{}'", corpus, searchString, maxRecords); |
---|
59 | |
---|
60 | SRUSearchRetrieveRequest searchRequest = new SRUSearchRetrieveRequest(corpus.getEndpoint().getUrl()); |
---|
61 | searchRequest.setVersion(version); |
---|
62 | searchRequest.setMaximumRecords(maxRecords); |
---|
63 | // searchRequest.setRecordSchema( |
---|
64 | // corpus.getEndpoint().getProtocol().equals(FCSProtocolVersion.LEGACY) |
---|
65 | // ? ClarinFCSRecordData.LEGACY_RECORD_SCHEMA |
---|
66 | // : ClarinFCSRecordData.RECORD_SCHEMA); |
---|
67 | searchRequest.setQuery("\"" + searchString + "\""); |
---|
68 | searchRequest.setStartRecord(startRecord); |
---|
69 | if (corpus.getHandle() != null) { |
---|
70 | searchRequest.setExtraRequestData(SRUCQL.SEARCH_CORPUS_HANDLE_PARAMETER, corpus.getHandle()); |
---|
71 | } |
---|
72 | requests.add(request); |
---|
73 | |
---|
74 | try { |
---|
75 | searchClient.searchRetrieve(searchRequest, new ThrottledClient.SearchCallback() { |
---|
76 | @Override |
---|
77 | public void onSuccess(SRUSearchRetrieveResponse response, ThrottledClient.Stats stats) { |
---|
78 | statistics.addEndpointDatapoint(corpus.getInstitution(), corpus.getEndpoint().getUrl(), stats.getQueueTime(), stats.getExecutionTime()); |
---|
79 | results.add(new Result(request, response, null)); |
---|
80 | requests.remove(request); |
---|
81 | } |
---|
82 | |
---|
83 | @Override |
---|
84 | public void onError(SRUSearchRetrieveRequest srureq, SRUClientException xc, ThrottledClient.Stats stats) { |
---|
85 | statistics.addEndpointDatapoint(corpus.getInstitution(), corpus.getEndpoint().getUrl(), stats.getQueueTime(), stats.getExecutionTime()); |
---|
86 | statistics.addErrorDatapoint(corpus.getInstitution(), corpus.getEndpoint().getUrl(), xc); |
---|
87 | results.add(new Result(request, null, xc)); |
---|
88 | requests.remove(request); |
---|
89 | } |
---|
90 | }); |
---|
91 | } catch (Throwable xc) { |
---|
92 | log.error("SearchRetrieve error for " + corpus.getEndpoint().getUrl(), xc); |
---|
93 | } |
---|
94 | return request; |
---|
95 | } |
---|
96 | |
---|
97 | public Long getId() { |
---|
98 | return id; |
---|
99 | } |
---|
100 | |
---|
101 | public List<Request> getRequests() { |
---|
102 | List<Request> copy = new ArrayList<>(); |
---|
103 | synchronized (requests) { |
---|
104 | copy.addAll(requests); |
---|
105 | } |
---|
106 | return copy; |
---|
107 | } |
---|
108 | |
---|
109 | public List<Result> getResults() { |
---|
110 | List<Result> copy = new ArrayList<>(); |
---|
111 | synchronized (results) { |
---|
112 | copy.addAll(results); |
---|
113 | } |
---|
114 | return copy; |
---|
115 | } |
---|
116 | |
---|
117 | public Statistics getStatistics() { |
---|
118 | return statistics; |
---|
119 | } |
---|
120 | |
---|
121 | public void exportTCF(TokenizerModel tokenizerModel) throws ExportException { |
---|
122 | byte[] bytes = Exports.getExportTokenizedTCF(results, searchLanguage, tokenizerModel); |
---|
123 | if (bytes != null) { |
---|
124 | Filedownload.save(bytes, "text/tcf+xml", "ClarinDFederatedContentSearch.xml"); |
---|
125 | } |
---|
126 | } |
---|
127 | |
---|
128 | public void exportText() { |
---|
129 | String text = Exports.getExportText(results); |
---|
130 | if (text != null) { |
---|
131 | Filedownload.save(text, "text/plain", "ClarinDFederatedContentSearch.txt"); |
---|
132 | } |
---|
133 | } |
---|
134 | |
---|
135 | void exportExcel() throws ExportException { |
---|
136 | byte[] bytes = Exports.getExportExcel(results); |
---|
137 | if (bytes != null) { |
---|
138 | Filedownload.save(bytes, "text/tcf+xml", "ClarinDFederatedContentSearch.xls"); |
---|
139 | } |
---|
140 | } |
---|
141 | |
---|
142 | void exportPWText(String user, String pass) { |
---|
143 | byte[] bytes = null; |
---|
144 | try { |
---|
145 | String text = Exports.getExportText(results); |
---|
146 | if (text != null) { |
---|
147 | bytes = text.getBytes(SEARCH_RESULTS_ENCODING); |
---|
148 | } |
---|
149 | } catch (Exception ex) { |
---|
150 | Logger.getLogger(Search.class.getName()).log(Level.SEVERE, null, ex); |
---|
151 | } |
---|
152 | if (bytes != null) { |
---|
153 | DataTransfer.uploadToPW(user, pass, bytes, "text/plan", ".txt"); |
---|
154 | } |
---|
155 | } |
---|
156 | |
---|
157 | String useWebLichtOnText() { |
---|
158 | String url = null; |
---|
159 | try { |
---|
160 | String text = Exports.getExportText(results); |
---|
161 | if (text != null) { |
---|
162 | byte[] bytes = text.getBytes(SEARCH_RESULTS_ENCODING); |
---|
163 | url = DataTransfer.uploadToDropOff(bytes, "text/plan", ".txt"); |
---|
164 | } |
---|
165 | } catch (Exception ex) { |
---|
166 | Logger.getLogger(Search.class.getName()).log(Level.SEVERE, null, ex); |
---|
167 | } |
---|
168 | return url; |
---|
169 | } |
---|
170 | |
---|
171 | String useWebLichtOnToks(TokenizerModel tokenizerModel) throws ExportException { |
---|
172 | String url = null; |
---|
173 | byte[] bytes = Exports.getExportTokenizedTCF(results, searchLanguage, tokenizerModel); |
---|
174 | if (bytes != null) { |
---|
175 | url = DataTransfer.uploadToDropOff(bytes, "text/tcf+xml", ".tcf"); |
---|
176 | } |
---|
177 | return url; |
---|
178 | } |
---|
179 | |
---|
180 | void exportPWExcel(String user, String pass) throws ExportException { |
---|
181 | byte[] bytes = Exports.getExportExcel(results); |
---|
182 | if (bytes != null) { |
---|
183 | DataTransfer.uploadToPW(user, pass, bytes, "application/vnd.ms-excel", ".xls"); |
---|
184 | } |
---|
185 | } |
---|
186 | |
---|
187 | public void exportPWTCF(String user, String pass, TokenizerModel tokenizerModel) throws ExportException { |
---|
188 | byte[] bytes = Exports.getExportTokenizedTCF(results, searchLanguage, tokenizerModel); |
---|
189 | if (bytes != null) { |
---|
190 | DataTransfer.uploadToPW(user, pass, bytes, "text/tcf+xml", ".tcf"); |
---|
191 | } |
---|
192 | } |
---|
193 | |
---|
194 | public void exportCSV() { |
---|
195 | String csv = Exports.getExportCSV(results, ";"); |
---|
196 | if (csv != null) { |
---|
197 | Filedownload.save(csv, "text/plain", "ClarinDFederatedContentSearch.csv"); |
---|
198 | } |
---|
199 | } |
---|
200 | |
---|
201 | public void exportPWCSV(String user, String pass) { |
---|
202 | String csv = Exports.getExportCSV(results, ";"); |
---|
203 | if (csv != null) { |
---|
204 | DataTransfer.uploadToPW(user, pass, csv.getBytes(), "text/csv", ".csv"); |
---|
205 | } |
---|
206 | } |
---|
207 | |
---|
208 | public void shutdown() { |
---|
209 | // nothing to do |
---|
210 | } |
---|
211 | |
---|
212 | public long getCreatedAt() { |
---|
213 | return createdAt; |
---|
214 | } |
---|
215 | } |
---|