public static void main(String[] args) throws ParseException, SAXException, TikaException {
LuceneTester tester;
try {
tester = new LuceneTester();
WriteFile = new WriteFile();
Scanner scanner = new Scanner(System.in);
// 이 machine의 JVM heap size를 출력. indexing을 하다가 OutOfMemoryError가 발생할 수 있기 때문.
long heapSize = Runtime.getRuntime().totalMemory() / (1024 * 1024);
System.out.println("Heap Size : " + heapSize + "MB");
System.out.print("Index(y/n):");
String IndexOrNot = scanner.nextLine();
if (IndexOrNot.compareTo("y") == 0) {
tester.createIndex();
}
scanner.close();
// Setup tcpip server for query from C# front-end
tcpIpServer = new TcpIpServer();
tcpIpServer.setup(LuceneConstants.PORT);
tcpIpServer.accept();
// 쿼리가 하나 올때마다 해당하는 검색을 수행하는 루프.
while (true) {
String input = ""; // Query from c#
ForSend_string = "[Search Result]\r\n"; // 실제적으로 사용하지는 아니하나, 디버깅과 검색이 완료되었다는 큐를 주기 위해 유지
ForSend_json = new JSONObject();
System.out.print("search query: ");
input = tcpIpServer.read();
// input = input.substring(1);
System.out.println(input);
/** ******메인 검색시에 수행 되는 서치************************ */
if ((input.substring(0, 1).compareTo("m")) == 0) {
// System.out.println("Main Search!" + input.substring(0, 1) +" " + input.substring(1));
tester.search(input.substring(1)); // search 수행
}
/** ******서브 검색시에 수행되는 서치************************ */
if ((input.substring(0, 1).compareTo("s")) == 0) {
// System.out.println("Sub Search!" + input.substring(0, 1) +" " + input.substring(1));
tester.sub_search(input.substring(1)); // sub-search 수행
}
// 검색 결과를 out.txt에 json format으로 저장. tcp전송은 큐를 날려주는 용도
WriteFile.write(ForSend_json.toJSONString());
tcpIpServer.write(ForSend_string);
}
} catch (IOException e) {
e.printStackTrace();
}
}
/** Sets the comment with the input from customizer. */
public void evaluateInput() {
writeFile.setOutput(jTextField1.getText());
writeFile.setOutputFile(jTextField2.getText());
writeFile.setComment(jTextField3.getText());
}
public static void main(String[] args) throws IOException {
try {
String type = "rawWords";
Utility u = new Utility(type);
String folder = u.returnFolderName();
Double avg_doc_length = u.returnAvgLength();
Map<String, DocBean> docCatBean = u.getDocCat();
Map<String, TokenCatalogBean> tokenCatBean = u.getTokenCat();
System.out.println("Avg Doc Length:: " + avg_doc_length);
System.out.println("DocCatBean Size:: " + docCatBean.size());
System.out.println("Token Cat Bean:: " + tokenCatBean.size());
System.out.println("Doc Bean:::: " + u.getDocBean().size());
long vocabSize = tokenCatBean.size();
System.out.println("Vocab Size ::" + vocabSize);
System.out.println("Avg Doc Length:::: " + avg_doc_length);
/* Method to read the query file */
String query_file_path =
"C:/Users/Nitin/NEU/Summer Sem/IR/Data/Assign 1/AP89_DATA/AP_DATA/query_desc.51-100.short.txt";
/* This will be later replaced by path from the config file */
GetFinalQueries qu = new GetFinalQueries();
List<String> queries = qu.readQueryFile(query_file_path);
/*
* Method to get stopwords from the file and append the common words
* from query file
*/
/* Changed the split regex from space to space and hypen */
List<String> stop_words_final = qu.getStopWords();
/*
* List<String> stop_words_final = new ArrayList<String>();
*
* for (int i = 0; i < stop_words_custom.length; i++) {
* stop_words_final.add(stop_words_custom[i]); }
*/
/*
* Method to remove stopwords from query and just get the final
* query
*/
/*
* Iterating the queries one by one. Each Query is a list of String
* (Query Words)
*/
List<List<String>> final_query = new ArrayList<List<String>>();
final_query = qu.getFinalQueryList(queries, stop_words_final);
/*
* for (List<String> query : final_query) {
* System.out.println("___________________________"); for (String q
* : query) { System.out.println(q); }
* System.out.println("___________________________"); }
*/
List<List<String>> resultOkapi = new ArrayList<List<String>>();
/* for (String query : queries) { */
for (List<String> query : final_query) {
Map<String, String> queryTFList = new HashMap<String, String>();
Map<String, Double> rankTerm = new HashMap<String, Double>();
String querynum = null;
System.out.println("Query Minus stop words");
System.out.println("=======================");
// System.out.println(query.get(0));
/*
* For every word in a query calculates the okapif value and
* sums it up
*/
querynum = query.get(0).replace(".", "");
// System.out.println("Query Numm::: "+ querynum);
System.out.println("Query Size::" + query.size());
for (int i = 1; i < query.size(); i++) {
/* Method to calculate tfs for each term in query */
String w = query.get(i).toLowerCase();
Map<String, Integer> tfMap = new HashMap<String, Integer>();
// System.out.println("Calculating for Word::: " + q);
tfMap =
laplaceSmoothing(
w.replaceAll("[,\"()]", ""),
avg_doc_length,
tokenCatBean,
docCatBean,
u,
vocabSize,
folder);
System.out.println(
"Size of TF Results:: " + tfMap.size() + "for :" + w.replaceAll("[,\"()]", ""));
for (Map.Entry<String, Integer> term : tfMap.entrySet()) {
if (queryTFList.get(term.getKey()) == null) {
queryTFList.put(term.getKey(), term.getValue().toString());
} else {
queryTFList.put(
term.getKey(), queryTFList.get(term.getKey()) + " " + term.getValue().toString());
// System.out.println("TF:: "+queryTFList.get(term.getKey()));
}
}
}
System.out.println("Final DOc List Size::: " + queryTFList.size());
System.out.println("Calculating Laplace Smoothing Score for each ::::::: ");
for (Map.Entry<String, String> d : queryTFList.entrySet()) {
double docLen = getDocLength(d.getKey(), docCatBean);
rankTerm.put(
d.getKey(),
laplacePerTerm(d.getValue(), docLen, avg_doc_length, vocabSize, query.size()));
}
/* Method to Sort Hashmap based on the value */
SortMap sm = new SortMap();
LinkedHashMap<String, Double> sortedRanks =
(LinkedHashMap<String, Double>) sm.getSortedRankMap(rankTerm);
int j = 1;
List<String> queryResults = new ArrayList<String>();
for (Entry<String, Double> term : sortedRanks.entrySet()) {
if (j <= 1000) {
String toWrite =
querynum
+ " "
+ "Q0"
+ " "
+ term.getKey()
+ " "
+ j
+ " "
+ term.getValue()
+ " "
+ "EXP";
// System.out.println(toWrite);
queryResults.add(toWrite);
} else {
// bw.newLine();
break;
}
j++;
}
resultOkapi.add(queryResults);
}
WriteFile w = new WriteFile();
w.writeToFile(resultOkapi, "Laplace-1.txt", type);
// node.close();
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
