本文整理汇总了Java中java.sql.ResultSetMetaData.getTableName方法的典型用法代码示例。如果您正苦于以下问题:Java ResultSetMetaData.getTableName方法的具体用法?Java ResultSetMetaData.getTableName怎么用?Java ResultSetMetaData.getTableName使用的例子?那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类java.sql.ResultSetMetaData
的用法示例。
在下文中一共展示了ResultSetMetaData.getTableName方法的7个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: getTableColName
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
private String getTableColName(int nColSourceIndex)
{
// DB2 JDBC Driver supports rsMetaData.getTableName(nColSourceIndex); See http://publib.boulder.ibm.com/infocenter/db2help/index.jsp?topic=/com.ibm.db2.udb.doc/ad/rjvjdapi.htm
try
{
ResultSetMetaData rsMetaData = m_r.getMetaData();
String csTableName = rsMetaData.getTableName(nColSourceIndex);
String csColName = rsMetaData.getColumnName(nColSourceIndex);
String csTableColName = SemanticContextDef.getTableColName(csTableName, csColName);
return csTableColName;
}
catch (SQLException e)
{
LogSQLException.log(e);
}
return "";
}
示例2: createTable
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
protected void createTable(){
try {
//Quoter qt = SQLIdentifiers.createQuoter(dbmd);
ResultSet rs = conn.createStatement().executeQuery(context.getSqlSelect());
ResultSetMetaData rsMeta = rs.getMetaData();
String aName = rsMeta.getTableName(1);
String aSchema = rsMeta.getSchemaName(1);
String aCatalog = rsMeta.getCatalogName(1);
table = new DBTable(aName, aSchema, aCatalog);
table.setQuoter(SQLIdentifiers.createQuoter(conn.getMetaData()));
//table.setQuoter(quoter);
} catch (SQLException ex) {
Exceptions.printStackTrace(ex);
}
}
示例3: createTable
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
protected void createTable() {
try {
//Quoter qt = SQLIdentifiers.createQuoter(dbmd);
ResultSet rs = conn.createStatement().executeQuery(context.getSqlSelect());
ResultSetMetaData rsMeta = rs.getMetaData();
String aName = rsMeta.getTableName(1);
String aSchema = rsMeta.getSchemaName(1);
String aCatalog = rsMeta.getCatalogName(1);
table = new DBTable(aName, aSchema, aCatalog);
//table.setQuoter(quoter);
} catch (SQLException ex) {
Exceptions.printStackTrace(ex);
}
}
示例4: ResultSetIterator
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
public ResultSetIterator(ResultSet resultSet, String relationName) throws SQLException {
this.resultSet = resultSet;
ResultSetMetaData resultSetMetaData = resultSet.getMetaData();
this.numberOfColumns = resultSetMetaData.getColumnCount();
this.nextCalled = false;
this.relationName = resultSetMetaData.getTableName(1);
if (this.relationName == null || this.relationName.isEmpty())
this.relationName = relationName; // use as fallback only
if (this.relationName == null || this.relationName.isEmpty())
this.relationName = UNKNOWN_RELATION_NAME;
this.columnNames = retrieveColumnNames(resultSetMetaData);
}
示例5: QueryResult
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
public QueryResult(ResultSet resultSet) {
if (resultSet == null) {
return;
}
try {
ResultSetMetaData md = resultSet.getMetaData();
for (int i = 0; i < md.getColumnCount(); i++) {
int index = i + 1;
VariableDescriptor desc = new VariableDescriptor(
md.getColumnName(index),
md.getColumnLabel(index),
md.getTableName(index)
);
variableDescriptors.add(desc);
}
while (resultSet.next()) {
List<Object> row = new ArrayList<>();
for (int i = 0; i < md.getColumnCount(); i++) {
Object value = resultSet.getObject(i + 1);
row.add(value);
}
rows.add(new DataRow(variableDescriptors, row));
}
} catch (Exception e) {
throw new RuntimeException(e);
}
}
示例6: getTableName
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
protected String getTableName(PredicateIndicator pi) throws SQLException {
ResultSetMetaData meta = belsDB.get(pi);
return meta.getTableName(1);
}
示例7: getMeta
import java.sql.ResultSetMetaData; //导入方法依赖的package包/类
/**
* Creates a new {@link MetaData} object for a given sql query.
* At this time only one table in the from clause can be specified
* due to a Oracle inefficiency in their JDBC driver meta data fetch.
* @param sql
* @param sourceConn
* @return
* @throws SQLException
* @throws IOException
*/
public static MetaData getMeta(String sql, Connection sourceConn)
throws SQLException, IOException {
PreparedStatement pstmt = null;
ResultSet rs = null;
try {
// To download the entire result set it is supposedly much
// faster to do forward only. This means that we can only
// call rs.next() and not rs.first(). So we must make sure
// that rs.next() is not called before we start downloading
// the table. This is called in populateTable.
pstmt = sourceConn.prepareStatement(sql,
ResultSet.TYPE_FORWARD_ONLY, ResultSet.CONCUR_READ_ONLY);
pstmt.setFetchSize(1);
pstmt.setMaxRows(1);
rs = pstmt.executeQuery();
// If no data was found quit.
if (rs == null) {
log.warn("No data found in <SOURCE> using sql statement:" + sql);
return null;
}
log.log(XLevel.DEBUG2, "Processing <SOURCE> data...");
// Get the table's metadata
ResultSetMetaData meta = rs.getMetaData();
// For the result metadata set, we need to cache the count and
// column names. Calls to getXXX for meta data can make calls
// back to the database. We're pretty sure they're not going to
// change while were running.
int cols = meta.getColumnCount() + 1;
ArrayList<String> sourceCols = new ArrayList<String>();
for (int i = 1; i < cols; i++)
sourceCols.add(meta.getColumnName(i));
// XXX Oracle 10 Sucks!
// Oracle does not implement the getTableName and getSchemaName
// methods. Since they do not implement them, we have to parse
// the query to determine the table name.
String tableName = meta.getTableName(1);
if (tableName.trim().length() == 0) {
Pattern p = Pattern.compile(".*\\s+from\\s+\\w*?\\.?(\\w+).*",
Pattern.CASE_INSENSITIVE);
Matcher m = p.matcher(sql);
if (m.matches())
tableName = m.group(1).toUpperCase();
}
MetaData metaData = new MetaData(tableName, meta, sourceConn.getMetaData());
return metaData;
} finally {
try {
if (rs != null)
rs.close();
if (pstmt != null)
pstmt.close();
} catch (SQLException ignored) {
}
}
}