searchCollector (use in limited conf)

build.xml

@@ -38,7 +38,7 @@
 
 	<target name="copy" depends="publish">
 		<echo>Copy to ICS6</echo>
-		<copy file="publish/isearcher_${version.number}.${build.number}.jar" todir="../../workspace/ICS6/ics/WEB-INF/lib/" />
+		<copy file="publish/isearcher_${version.number}.${build.number}.jar" todir="../ICS6_WORK/ics/WEB-INF/lib/" />
 		<echo>Copy to ICS6Extend</echo>
 		<copy file="publish/isearcher_${version.number}.${build.number}.jar" todir="../../workspace/ICS6Extend/lib/common/" />
 		<echo>Copy to AradonExtend</echo>

src/net/ion/nsearcher/common/WriteDocument.java

@@ -43,19 +43,27 @@ public class WriteDocument extends AbDocument {
 	private IndexSession isession;
 	private boolean newDoc = false;
 	private float boost = 1.0f ;
+	private Document doc;
 
 	public WriteDocument(IndexSession indexSession, String docId) {
 		this.isession = indexSession ;
 		this.docId = docId;
+		this.doc = new Document();
+	}
+
+	public WriteDocument(IndexSession indexSession, String docId, Document doc) {
+		this.isession = indexSession ;
+		this.docId = docId;
+		this.doc = (doc == null) ? new Document() : doc;
 	}
 
 	public WriteDocument(IndexSession indexSession) {
 		this.isession = indexSession ;
 		this.docId = new ObjectId().toString();
 		this.newDoc = true ;
+		this.doc = new Document();
 	}
 
-
 	public String idValue() {
 		return docId;
 	}
@@ -72,7 +80,6 @@ public WriteDocument boost(float boost){
 	public Document toLuceneDoc() {
 
 		FieldIndexingStrategy strategy = isession.fieldIndexingStrategy(); 
-		Document doc = new Document();
 		StringBuilder bodyBuilder = new StringBuilder(512);
 		bodyBuilder.append(docId + " ") ;
 
@@ -194,6 +201,12 @@ public WriteDocument text(String fieldName, String value) {
 		add(MyField.text(fieldName, value));
 		return this;
 	}
+
+	public WriteDocument stext(String fieldName, String value) {
+		add(MyField.manual(fieldName, value, Store.YES, true, MyFieldType.Text));
+		return this;
+	}
+
 
 	public WriteDocument number(String fieldName, long value) {
 		add(MyField.number(fieldName, value));
@@ -256,4 +269,5 @@ public Void insertVoid() throws IOException {
 		return null ;
 	}
 
+
 }

src/net/ion/nsearcher/index/IndexSession.java

@@ -5,6 +5,7 @@
 import net.ion.framework.util.MapUtil;
 import net.ion.nsearcher.common.AbDocument.Action;
 import net.ion.nsearcher.common.FieldIndexingStrategy;
+import net.ion.nsearcher.common.MyField;
 import net.ion.nsearcher.common.SearchConstant;
 import net.ion.nsearcher.common.WriteDocument;
 import net.ion.nsearcher.search.SingleSearcher;
@@ -16,6 +17,7 @@
 import org.apache.lucene.index.IndexWriter;
 import org.apache.lucene.index.IndexWriterConfig;
 import org.apache.lucene.index.Term;
+import org.apache.lucene.queryparser.classic.ParseException;
 import org.apache.lucene.search.Query;
 import org.apache.lucene.store.Directory;
 
@@ -72,6 +74,14 @@ public WriteDocument newDocument(){
 		return new WriteDocument(this) ;
 	}
 
+
+	public WriteDocument loadDocument(String docId) throws IOException, ParseException {
+		Document findDoc = searcher.central().newSearcher().createRequestByKey(docId).findOne().toLuceneDoc() ;
+		WriteDocument result = new WriteDocument(this, docId, findDoc);
+		return result;
+	}
+
+
 	public FieldIndexingStrategy fieldIndexingStrategy() {
 		return fieldIndexingStrategy;
 	}

src/net/ion/nsearcher/search/AbstractDocCollector.java

@@ -0,0 +1,26 @@
+package net.ion.nsearcher.search;
+
+import java.io.IOException;
+import java.util.Set;
+
+import org.apache.lucene.index.DirectoryReader;
+
+import net.ion.nsearcher.common.ReadDocument;
+
+public abstract class AbstractDocCollector implements DocCollector {
+	public abstract ColResult accept(ReadDocument doc) ;
+
+	public ColResult accept(DirectoryReader dreader, SearchRequest sreq, int docId) throws IOException{
+		return accept(toDoc(dreader, sreq, docId)) ;
+	}
+
+	private ReadDocument toDoc(DirectoryReader dreader, SearchRequest sreq, int docId) throws IOException {
+		Set<String> fields = sreq.selectorField();
+		if (fields == null || fields.size() == 0) {
+			return ReadDocument.loadDocument(dreader.document(docId));
+		}
+		return ReadDocument.loadDocument(dreader.document(docId, sreq.selectorField()));
+	}
+
+
+}

src/net/ion/nsearcher/search/DocCollector.java

@@ -0,0 +1,20 @@
+package net.ion.nsearcher.search;
+
+import java.io.IOException;
+
+import org.apache.lucene.index.DirectoryReader;
+
+public interface DocCollector {
+
+	public enum ColResult {
+		ACCEPT, REVOKE, BREAK
+	}
+
+	public final static DocCollector BLANK = new DocCollector() {
+		public ColResult accept(DirectoryReader dreader, SearchRequest sreq, int docId) {
+			return ColResult.ACCEPT;
+		}
+	};
+
+	public ColResult accept(DirectoryReader dreader, SearchRequest sreq, int docId) throws IOException ;
+}

src/net/ion/nsearcher/search/PageCollector.java

@@ -0,0 +1,32 @@
+package net.ion.nsearcher.search;
+
+import java.io.IOException;
+
+import net.ion.framework.db.Page;
+
+import org.apache.lucene.index.DirectoryReader;
+
+public class PageCollector implements DocCollector{
+
+	private Page page ;
+	private DocCollector pre;
+	private int count = -1 ;
+
+	public PageCollector(Page page, DocCollector pre){
+		this.page = page ;
+		this.pre  = pre  ;
+	}
+
+	@Override
+	public ColResult accept(DirectoryReader dreader, SearchRequest sreq, int docId) throws IOException {
+		if (pre.accept(dreader, sreq, docId) == ColResult.ACCEPT){
+			count++ ;
+			if (count >= page.getStartLoc() && count < page.getEndLoc()){
+				return ColResult.ACCEPT ;
+			} else if (count == page.getEndLoc()){
+				return ColResult.BREAK ;
+			}
+		}
+		return ColResult.REVOKE ;
+	}
+}

src/net/ion/nsearcher/search/SearchRequest.java

@@ -38,6 +38,7 @@ public class SearchRequest {
 	private Set<String> columns = SetUtil.newSet() ;
 	private Set<String> lazyColumns = SetUtil.newSet() ;
 	private String userDefine = "";
+	private DocCollector collector = DocCollector.BLANK;
 
 	SearchRequest(Searcher searcher, Query query){
 		this.searcher = searcher ;
@@ -229,5 +230,14 @@ public StoredFieldVisitor selector(){
 		return columns.size() == 0 ? null : new DocumentStoredFieldVisitor(columns) ;
 	}
 
+	public SearchRequest collect(DocCollector collector) {
+		this.collector = collector ;
+		return this;
+	}
+
+	public DocCollector collector(){
+		return collector ;
+	}
+
 
 }

src/net/ion/nsearcher/search/SearchResponse.java

@@ -24,22 +24,26 @@ public class SearchResponse {
 	private final long endTime;
 	private Future<Void> postFuture ;
 	private List<Integer> docs ;
-	private TopDocs tdocs;
-	private SearchResponse(ISearchable searcher, SearchRequest sreq, List<Integer> docs, TopDocs tdocs, long startTime) {
+	private int totalCount ;
+	private SearchResponse(ISearchable searcher, SearchRequest sreq, List<Integer> docs, int totalCount, long startTime) {
 		this.searcher = searcher ;
 		this.sreq = sreq ;
 		this.startTime = startTime;
 		this.endTime = System.currentTimeMillis();
 		this.docs = docs ;
-		this.tdocs = tdocs ;
+		this.totalCount = totalCount ;
 	}
 
+	public static SearchResponse create(ISearchable searcher, SearchRequest sreq, List<Integer> docs, int totalCount, long startTime) throws IOException {
+		return new SearchResponse(searcher, sreq, makeDocument(sreq, docs), totalCount, startTime) ;
+	}
+
 	public static SearchResponse create(ISearchable searcher, SearchRequest sreq, TopDocs docs, long startTime) throws IOException {
-		return new SearchResponse(searcher, sreq, makeDocument(searcher, sreq, docs), docs, startTime);
+		return new SearchResponse(searcher, sreq, makeDocument(sreq, docs), docs.totalHits, startTime);
 	}
 
 	public int totalCount() {
-		return tdocs.totalHits ;
+		return totalCount ;
 		// 전체 total은 searcherImpl이 구함. 
 //		return searcher.totalCount(sreq, sreq.getFilter()) ;
 	}
@@ -87,7 +91,16 @@ public List<ReadDocument> getDocument() throws IOException{
 		return eachDoc(EachDocHandler.TOLIST) ;
 	}
 
-	private static List<Integer> makeDocument(ISearchable searcher, SearchRequest sreq, TopDocs docs) {
+	private static List<Integer> makeDocument(SearchRequest sreq, List<Integer> docs) {
+		List<Integer> result = ListUtil.newList() ;
+
+		for (int i = sreq.skip(); i < Math.min(sreq.limit(), docs.size()); i++) {
+			result.add(docs.get(i));
+		}
+		return result;
+	}
+
+	private static List<Integer> makeDocument(SearchRequest sreq, TopDocs docs) {
 		ScoreDoc[] sdocs = docs.scoreDocs;
 		List<Integer> result = ListUtil.newList() ;
 

src/net/ion/nsearcher/search/SingleSearcher.java

@@ -2,28 +2,36 @@
 
 import java.io.Closeable;
 import java.io.IOException;
+import java.util.List;
 import java.util.Set;
 import java.util.concurrent.Callable;
 import java.util.concurrent.ExecutorService;
 import java.util.concurrent.Executors;
 import java.util.concurrent.Future;
 import java.util.concurrent.TimeUnit;
+import java.util.concurrent.atomic.AtomicInteger;
 import java.util.concurrent.locks.Lock;
 
 import net.ion.framework.util.Debug;
+import net.ion.framework.util.ListUtil;
 import net.ion.nsearcher.common.IKeywordField;
 import net.ion.nsearcher.common.ReadDocument;
 import net.ion.nsearcher.config.Central;
 import net.ion.nsearcher.config.SearchConfig;
 import net.ion.nsearcher.reader.InfoReader;
+import net.ion.nsearcher.search.DocCollector.ColResult;
 
 import org.apache.lucene.document.Document;
+import org.apache.lucene.index.AtomicReaderContext;
 import org.apache.lucene.index.DirectoryReader;
 import org.apache.lucene.index.IndexReader;
 import org.apache.lucene.index.Term;
+import org.apache.lucene.search.Collector;
 import org.apache.lucene.search.Filter;
 import org.apache.lucene.search.IndexSearcher;
+import org.apache.lucene.search.MatchAllDocsQuery;
 import org.apache.lucene.search.ScoreDoc;
+import org.apache.lucene.search.Scorer;
 import org.apache.lucene.search.Sort;
 import org.apache.lucene.search.SortField;
 import org.apache.lucene.search.SortField.Type;
@@ -50,7 +58,7 @@ public static SingleSearcher create(SearchConfig sconfig, Central central) throw
 		return new SingleSearcher(central, DirectoryReader.open(central.dir()));
 	}
 
-	public SearchResponse search(SearchRequest sreq, Filter filters) throws IOException {
+	public SearchResponse search(final SearchRequest sreq, Filter filters) throws IOException {
 		reloadReader();
 
 		Lock rlock = central.readLock() ;
@@ -59,6 +67,33 @@ public SearchResponse search(SearchRequest sreq, Filter filters) throws IOExcept
 			locked = rlock.tryLock(); 
 			long startTime = System.currentTimeMillis();
 
+			if (sreq.collector() != AbstractDocCollector.BLANK){
+				final AtomicInteger total = new AtomicInteger() ;
+				final List<Integer> docs = ListUtil.newList() ;
+				try {
+					isearcher.search(sreq.query(), filters, new Collector() {
+						public void setScorer(Scorer scorer) throws IOException {
+						}
+						public void setNextReader(AtomicReaderContext atomicreadercontext) throws IOException {
+						}
+						public void collect(int docId) throws IOException {
+							ColResult cresult = sreq.collector().accept(dreader, sreq, docId);
+							if (cresult == ColResult.ACCEPT) {
+								docs.add(docId) ;
+								total.incrementAndGet() ;
+							}
+							if (cresult == ColResult.BREAK) throw new IllegalStateException("break") ;
+						}
+						public boolean acceptsDocsOutOfOrder() {
+							return false;
+						}
+					});
+				} catch(IllegalStateException ignore){
+				} 
+				return SearchResponse.create(this, sreq, docs, total.intValue(), startTime) ;
+			}
+
+
 			TopDocs docs = isearcher.search(sreq.query(), filters, sreq.limit(), sreq.sort());
 			return SearchResponse.create(this, sreq, docs, startTime);
 		} finally {
@@ -79,6 +114,8 @@ public Document findById(String id) throws IOException{
 	public int totalCount(SearchRequest sreq, Filter filters) {
 		try {
 			// reloadReader() ;
+			if (sreq.collector() == AbstractDocCollector.BLANK) throw new IllegalArgumentException("with collector condition, this method meanless") ;
+
 			TopDocs docs = isearcher.search(sreq.query(), filters, Integer.MAX_VALUE);
 			return docs.totalHits;
 		} catch (IOException e) {

src/org/apache/lucene/analysis/ko/MyKoreanAnalyzer.java

@@ -5,20 +5,12 @@
 import java.io.Reader;
 import java.io.StringReader;
 import java.nio.CharBuffer;
-import java.util.Arrays;
-import java.util.List;
-import java.util.Set;
 
 import net.ion.framework.util.IOUtil;
 
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.core.LowerCaseFilter;
 import org.apache.lucene.analysis.core.StopFilter;
-import org.apache.lucene.analysis.ko.HanjaMappingFilter;
-import org.apache.lucene.analysis.ko.KoreanFilter;
-import org.apache.lucene.analysis.ko.KoreanTokenizer;
-import org.apache.lucene.analysis.ko.PunctuationDelimitFilter;
-import org.apache.lucene.analysis.ko.WordSegmentFilter;
 import org.apache.lucene.analysis.standard.ClassicFilter;
 import org.apache.lucene.analysis.util.CharArraySet;
 import org.apache.lucene.analysis.util.StopwordAnalyzerBase;

test/net/ion/nsearcher/index/TestDocument.java

@@ -143,4 +143,29 @@ public Document handle(IndexSession isession) throws Exception {
 		request.find().debugPrint();
 	}
 
+
+	public void testLoadDocument() throws Exception {
+		Document doc = indexer.index(new IndexJob<Document>() {
+			public Document handle(IndexSession isession) throws Exception {
+				final WriteDocument writeDoc = isession.newDocument("bleujin").keyword("@path", "_emp").keyword("@path", "/ion").stext("explain", "hello bleujin");
+				isession.insertDocument(writeDoc);
+				return writeDoc.toLuceneDoc();
+			}
+		});
+
+//		cen.newSearcher().createRequestByKey("bleujin").find().debugPrint(); 
+
+		indexer.index(new IndexJob<Void>() {
+			@Override
+			public Void handle(IndexSession isession) throws Exception {
+				isession.loadDocument("bleujin").number("age", 20).update() ;
+				return null;
+			}
+		}) ;
+
+		ReadDocument rdoc = cen.newSearcher().createRequest("").findOne() ;
+		assertEquals("hello bleujin", rdoc.asString("explain"));
+		assertEquals(true, cen.newSearcher().createRequestByTerm("explain", "bleujin").findOne() != null) ;
+	}
+
 }