1.查找數(shù)據(jù)庫中表的列名
1
2
3
4
5
6
7
8
9
10
11
12
13
|
<pre name= "code" class = "html" >String sql = "select *from tblmetadatainfo" ; ResultSet rs = MySqlHelper.executeQuery(sql, null ); String str= "" ; try { ResultSetMetaData rsmd = rs.getMetaData(); for ( int i = 1 ; i < rsmd.getColumnCount(); i++) { str+=rsmd.getColumnName(i)+ "," ; } str=str.substring( 0 , str.length()- 1 ); } catch (SQLException e) { // TODO Auto-generated catch block e.printStackTrace(); } |
2.查找數(shù)據(jù)庫中表中每條記錄的列值
1
2
3
|
for ( int i= 1 ;i<rs.getMetaData().getColumnCount();i++){ str+=rs.getString(i)+ "," ; } |
補(bǔ)充知識(shí):Java:使用ResultSet.next()執(zhí)行含有rownum的SQL語句速度緩慢
在使用Oracle數(shù)據(jù)庫進(jìn)行分頁查詢時(shí),經(jīng)常會(huì)用到如下SQL:
select tm.* from (select rownum rm, t.* xmlrecord from testtable t) tm where tm.rm > ? and tm.rm <= ?
使用的java代碼如下:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
int startIdx = 0 ; int endIdx = 10000 ; String sql = "select tm.* from (select rownum rm, t.* xmlrecord from testtable t) tm where tm.rm > ? and tm.rm <= ?" ; try (Connection conn = dataSource.getConnection(); PreparedStatement ps = conn.prepareStatement(sql);) { ps.setInt( 1 , startIdx); ps.setInt( 2 , endIdx); try (ResultSet rs = ps.executeQuery();) { while (rs.next()) { String id = rs.getString( 2 ); System.out.println( "id=" +id); } } } |
當(dāng)使用以上代碼時(shí),會(huì)發(fā)現(xiàn)當(dāng)取完最后一條記錄后,再執(zhí)行rs.next()時(shí),本來希望返回false后跳出循環(huán),但rs.next()會(huì)執(zhí)行非常長的時(shí)間。解決的方法是不讓rs.next()來判斷下一條記錄不存在,而在代碼通過計(jì)數(shù)來實(shí)現(xiàn):
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
|
int startIdx = 0 ; int endIdx = 10000 ; int i = 0 ; int count = endIdx - startIdx; String sql = "select tm.* from (select rownum rm, t.* xmlrecord from testtable t) tm where tm.rm > ? and tm.rm <= ?" ; try (Connection conn = dataSource.getConnection(); PreparedStatement ps = conn.prepareStatement(sql);) { ps.setInt( 1 , startIdx); ps.setInt( 2 , endIdx); try (ResultSet rs = ps.executeQuery();) { while (rs.next()) { i++; String id = rs.getString( 2 ); System.out.println( "id=" +id); if (i == count) { break ; } } } } |
如果代碼中設(shè)置了fetchSize,并且fetchSize不能被count整除時(shí),在取最后一片數(shù)據(jù)時(shí),rs.next()也會(huì)執(zhí)行很長時(shí)間:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
|
int startIdx = 0 ; int endIdx = 10000 ; String sql = "select tm.* from (select rownum rm, t.* xmlrecord from testtable t) tm where tm.rm > ? and tm.rm <= ?" ; try (Connection conn = dataSource.getConnection(); PreparedStatement ps = conn.prepareStatement(sql);) { ps.setFetchSize( 300 ); ps.setInt( 1 , startIdx); ps.setInt( 2 , endIdx); try (ResultSet rs = ps.executeQuery();) { while (rs.next()) { String id = rs.getString( 2 ); System.out.println( "id=" +id); } } } |
以上代碼中,當(dāng)取得9900條數(shù)據(jù)后,再取下一個(gè)300條時(shí),rs.next()就會(huì)執(zhí)行很長時(shí)間,可能是由于取不到一個(gè)完整的300條記錄造成的。解決方法是將fetchSize設(shè)置成能被count整除的數(shù)字,比如:
ps.setFetchSize(500);
在以上兩種狀況下,為什么rs.next()會(huì)執(zhí)行很長時(shí)間,還不是很清楚,但可以通過上述方式解決。至于為什么會(huì)有這個(gè)問題,有知道原因的朋友,請(qǐng)不吝賜教。
以上這篇java中ResultSet遍歷數(shù)據(jù)操作就是小編分享給大家的全部內(nèi)容了,希望能給大家一個(gè)參考,也希望大家多多支持服務(wù)器之家。
原文鏈接:https://blog.csdn.net/liuyunshengsir/article/details/50618494