package index;
import gragh.makeGragh;
import java.io.IOException;
import java.util.Date;
import java.util.Map;
import java.util.HashMap;
import java.util.Iterator;
import database.dataAccess;
import javaToHdfs.FsDirectory;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.CorruptIndexException;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.util.Version;
import org.apache.lucene.document.Field;
public class MakeForm {
private static Map<String,String[]> FKmap = new HashMap<String,String[]>();
private static Map<String,String> Attrmap = new HashMap<String,String>();
private static Map<String,String[]> OperationQuery = new HashMap<String,String[]>();
private static Map<String,String> Graghmap = new HashMap<String,String>();
public static void Form()
{
dataAccess db = new dataAccess();
String table[] = db.getAllTableName2(); //获得数据库中表的名称
for(String t:table)
{
int AC = db.AbsoluteCardinality(t); //实体的绝对基数
String[] attr = db.getAllAttribute(t);//获得表的属性的名称
String attrtxt = "";
for(int i=0;i<attr.length;i++)
{
String[] Query = new String[2];
double NA = db.AttributeNecessity(attr[i], t, AC);
double AR = db.AttributeRange(attr[i], t, AC);
double QA_S = NA*AR; //基于选择操作的属性的查询意向度
Query[0] = QA_S+"";
int NR;
if(db.IsNum(t, attr[i])==true && AR<1)
NR = 1;
else
NR = 0;
double QA_A = NA*NR; //基于聚合操作的属性的查询意向度
double QA_AS = 0.5*QA_S + 0.5*QA_A;
Query[1] = QA_AS+"";
OperationQuery.put(attr[i], Query);
}
for(int i=0;i<attr.length-1;i++)
{
attrtxt += attr[i] + ",";
}
attrtxt += attr[attr.length-1];
Attrmap.put(t,attrtxt);
String FK[] = db.getAllFK(t); //获得数据库中表的名称
FKmap.put(t, FK);
}
Graghmap = makeGragh.Graph(table, FKmap);
db.close();
}
public static void indexDocs(IndexWriter writer,Map<String,String> Attrmap,Map<String,String> Graphmap)
throws CorruptIndexException, IOException
{
String[] sql = {"","avg","count","max","min","sum","group by","group by,avg",
"group by,count","group by,max","group by,min","group by,sum"};
//首先完成单表的表单
Iterator<String> iter1 = Attrmap.keySet().iterator();
String k,v,allv;
int e,j=0;
while(iter1.hasNext())
{
k = iter1.next();
v = Attrmap.get(k);
allv = v;
String[] attr = v.split(",");
if(attr.length > 6) //表单上显示的属性数目不应大于5个
{
//只进行选择操作
for(int i=0;i<attr.length-1;i++) //利用冒泡排序将分数按照Q_S排序
{
for(int w=0;w<attr.length-1-i;w++)
{
String[] Query1 = OperationQuery.get(attr[w]);
String[] Query2 = OperationQuery.get(attr[w+1]);
double s1 = Double.parseDouble(Query1[0]);
double s2 = Double.parseDouble(Query2[0]);
if(s1<s2)
{
String temp = attr[w];
attr[w] = attr[w+1];
attr[w+1] = temp;
}
}
}
v = "";
for(int i=0;i<5;i++)
{
//System.out.print(attr[i]+" ");
//String[] Query = OperationQuery.get(attr[i]);
//System.out.println(Query[0]);
v += attr[i] + ",";
}
v += attr[5];
}
for(int i=0;i<sql.length;i++)
{
String all = k+","+allv+","+sql[i];
String con = k+","+v+","+sql[i];
e = 12*j+(i+1);
String formID = "form_" + e;
String descriptionA = "show " + v + " from " + k ;
String descriptionB = "use '" + sql[i] + "' when do searching in the database";
try {
Document doc = new Document();
doc.add(new Field("formID",formID, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("descriptionA",descriptionA, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("descriptionB",descriptionB, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("table", k, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("attr", v, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("sql", sql[i], Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("content", con, Field.Store.YES, Field.Index.ANALYZED));
doc.add(new Field("all", all, Field.Store.YES, Field.Index.ANALYZED));
doc.add(new Field("type", "1", Field.Store.YES, Field.Index.NOT_ANALYZED));
System.out.print("formID:"+formID+" ");
System.out.print("content:"+con);
System.out.print("type:"+ "1");
System.out.println("");
writer.addDocument(doc);
} finally{}
}
j++;
}
//然后完成多表的表单
Iterator<String> iter2 = Graphmap.keySet().iterator();
while(iter2.hasNext())
{
String alla="",attrs="",table="",table_FK;
k = iter2.next();
table_FK = Graphmap.get(k);
String[] tables = k.split(",");
for(int i=0;i<tables.length-1;i++)
{
table += tables[i]+",";
String at = Attrmap.get(tables[i]);
attrs += at+",";
}
table += tables[tables.length-1];
attrs += Attrmap.get(tables[tables.length-1]);
alla = attrs;
String[] attr = attrs.split(",");
//System.out.println("************************************************");
//System.out.println("table: "+table);
//System.out.println("attr: "+attrs);
if(attr.length > 6) //表单上显示的属性数目不应大于5个
{
//只进行选择操作
for(int i=0;i<attr.length-1;i++) //利用冒泡排序将分数按照Q_S排序
{
for(int w=0;w<attr.length-1-i;w++)
{
String[] Query1 = OperationQuery.get(attr[w]);
String[] Query2 = OperationQuery.get(attr[w+1]);
double s1 = Double.parseDouble(Query1[0]);
double s2 = Double.parseDouble(Query2[0]);
if(s1<s2)
{
String temp = attr[w];
attr[w] = attr[w+1];
attr[w+1] = temp;
}
}
}
attrs = "";
for(int i=0;i<5;i++)
{
//System.out.print(attr[i]+" ");
//String[] Query = OperationQuery.get(attr[i]);
//System.out.println(Query[0]);
attrs += attr[i] + ",";
}
attrs += attr[5];
}
for(int i=0;i<sql.length;i++)
{
e = 12*j+(i+1);
String formID = "form_" + e;
String con = table + "," + attrs + "," + sql[i];
String all = table + "," + alla + "," + sql[i];
String descriptionA = "show " + attrs + " ...from " + table;
String descriptionB = "use '" + sql[i] + "' when do searching in the database";
try {
Document doc = new Document();
doc.add(new Field("formID",formID, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("descriptionA",descriptionA, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("descriptionB",descriptionB, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("table",table, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("attr", attrs, Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("sql", sql[i], Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("content", con, Field.Store.YES, Field.Index.ANALYZED));
doc.add(new Field("all", all, Field.Store.YES, Field.Index.ANALYZED));
doc.add(new Field("type", "2", Field.Store.YES, Field.Index.NOT_ANALYZED));
doc.add(new Field("FK", table_FK , Field.Store.YES, Field.Index.NOT_ANALYZED));
System.out.print("formID:"+formID+" ");
System.out.print("content:"+con);
System.out.print("type:"+ "2");
System.out.println("");
writer.addDocument(doc);
} finally{}
}
j++;
}
}
public static void main(String[] args)
{
Date start = new Date();
String indexPath = "hdfs://localhost:9000/user/root/FormIndex";
try {
Configuration conf = new Configuration();
FileSyst
人工智能-项目实践-信息检索-关系数据库查询界面检索系统 （138个子文件） 
MakeForm.class 10KB makeGragh.class 8KB FsDirectory.class 6KB indexFiles.class 5KB dataAccess.class 5KB LuceneWriter.class 5KB LuceneOnHdfsSearch2.class 4KB DataIndex.class 4KB LuceneOnHdfsSearch.class 3KB IndexFiles2.class 3KB Test.class 3KB indexFiles3.class 3KB indexFiles3$Map.class 3KB indexFiles3$Reduce.class 3KB IndexFiles2$Reduce.class 3KB searchFiles.class 3KB FsDirectory$DfsIndexInput.class 2KB Enumerate.class 2KB FsDirectory$DfsIndexOutput.class 2KB dataAccessBase.class 2KB IndexFiles2$Map.class 2KB HDFSDocument.class 2KB FormIndex.class 2KB HDFSDocumentOutPut$1.class 2KB HDFSDocumentOutPut.class 2KB FsDirectory$1.class 1KB HadoopFSUtil.class 1KB MakeForm2.class 1KB FsDirectory$DfsIndexInput$Descriptor.class 1KB HadoopFSUtil$2.class 1KB Node.class 1024B Link.class 862B HadoopFSUtil$1.class 627B IsNum.class 545B .classpath 828B org.eclipse.wst.common.component 426B org.eclipse.wst.jsdt.ui.superType.container 49B
closed1.gif 85B opened1.gif 83B child1.gif 48B
aspectjtools-1.6.11.jar 8.51MB hadoop-core-1.2.1.jar 4.01MB core-3.1.1.jar 3.4MB mockito-all-1.8.5.jar 1.35MB lucene-core-3.0.1.jar 1002KB commons-math-2.1.jar 813KB mysql-connector-java-5.1.18-bin.jar 771KB hsqldb-1.8.0.10.jar 690KB jersey-server-1.8.jar 678KB jackson-mapper-asl-1.8.8.jar 653KB commons-collections-3.2.1.jar 562KB jetty-6.1.26.jar 527KB jersey-core-1.8.jar 447KB jasper-compiler-5.5.12.jar 396KB log4j-1.2.15.jar 383KB hadoop-tools-1.2.1.jar 377KB jets3t-0.6.1.jar 314KB commons-configuration-1.6.jar 292KB commons-httpclient-3.0.1.jar 273KB commons-net-3.1.jar 267KB commons-lang-2.4.jar 256KB jackson-core-asl-1.8.8.jar 222KB jdeb-0.8.jar 216KB commons-beanutils-core-1.8.0.jar 201KB junit-4.5.jar 194KB commons-beanutils-1.7.0.jar 184KB jsch-0.1.42.jar 181KB jetty-util-6.1.26.jar 173KB commons-io-2.1.jar 159KB jersey-json-1.8.jar 144KB commons-digester-1.8.jar 140KB aspectjrt-1.6.11.jar 113KB commons-el-1.0.jar 110KB jasper-runtime-5.5.12.jar 75KB hadoop-fairscheduler-1.2.1.jar 69KB oro-2.0.8.jar 64KB commons-logging-1.1.1.jar 59KB hadoop-capacity-scheduler-1.2.1.jar 57KB commons-codec-1.4.jar 57KB asm-3.2.jar 42KB commons-cli-1.2.jar 40KB commons-logging-api-1.0.4.jar 26KB slf4j-api-1.4.3.jar 15KB xmlenc-0.52.jar 15KB commons-daemon-1.0.1.jar 13KB kfs-0.2.2.jar 11KB hadoop-thriftfs-1.2.1.jar 10KB slf4j-log4j12-1.4.3.jar 8KB hadoop-ant-1.2.1.jar 7KB hadoop-minicluster-1.2.1.jar 417B hadoop-client-1.2.1.jar 414B MakeForm.java 9KB makeGragh.java 8KB FsDirectory.java 7KB dataAccess.java 5KB indexFiles.java 4KB indexFiles3.java 4KB LuceneOnHdfsSearch2.java 4KB IndexFiles2.java 3KB LuceneOnHdfsSearch.java 3KB共 138 条
- 1
- 2
资源评论
