我们正在使用EmbeddedId，我们在其中一个Entity中使用了Role:
+-----------------------------+
| roleId | name | discription |
+-----------------------------+
Rights:
+-----------------------------+
| rightId | name | discription|
+-----------------------------+
rightrole
+--------------------------------------+
| roleId | rightId | some other column |
+--------------------------------------+
。
表格如下：
@Entity
@Table(name = "role")
public class Role {
private static final long serialVersionUID = 1L;
@Id
@GeneratedValue(strategy = GenerationType.IDENTITY)
@Basic(optional = false)
@NotNull
@Column(nullable = false)
private Long roleID;
@OneToMany(cascade = CascadeType.ALL, mappedBy = "role", fetch = FetchType.LAZY)
private List<RightRole> rightRoleList;
.......
}
角色表的实体是：
@Entity
@Table(name = "rightrole")
public class RightRole extends BaseEntity<RightRolePK> {
private static final long serialVersionUID = 1L;
@EmbeddedId
protected RightRolePK rightRolePK;
@JoinColumn(name = "roleID", referencedColumnName = "roleID", insertable = false, updatable = false)
@ManyToOne(fetch = FetchType.LAZY)
private Role role;
@JoinColumn(name = "rightID", referencedColumnName = "rightID", insertable = false, updatable = false)
@ManyToOne(fetch = FetchType.LAZY)
private Right right;
......
}
@Embeddable
public class RightRolePK implements Serializable {
private static final long serialVersionUID = 1L;
@Basic(optional = false)
@NotNull
@Column(nullable = false)
private long roleID;
@Basic(optional = false)
@NotNull
@Column(nullable = false)
private long rightID;
.....
}
rightrole表的实体是：
role我的问题是每当我想用rights创建新的store(persist)时，首先我需要role flush个对象然后我必须id为角色获取新生成的rightrole。然后我可以把它放在rightrole实体的对象中。
有什么方法可以在角色对象中设置Auto Generated列表并一次性保留它。
这种冲洗套管是我们的性能瓶颈，因为对于批量插入，我们必须坚持单个单个物体。
我们正在使用 - (IBAction)selectBtn:(id)sender {
ParcelPickUpViewController *vc = [[HomeScreenVC alloc]initWithNibName:@"pickUpVC" bundle:nil];
[self.viewController pushViewController:vc animated: YES];
[self.viewController setNavigationBarHidden:YES animated:YES];
}
主键。
答案 0 :(得分：2)
JPA 2.0允许派生ID，通过添加@MapsId进行扩展以更好地支持这一点。保持其他一切但使用：
public static class MapForEquityCalculator extends Mapper<LongWritable, Text, Text, IntWritable> {
public int GetIndex(int R1, int R2, boolean IsSuited) {
if (IsSuited) {
return (int)(((13 - Math.ceil((double)Math.max(R1, R2) / 4)) * 13) + (13 - Math.ceil((double)Math.min(R1, R2) / 4)));
} else {
return (int)(((13 - Math.ceil((double)Math.min(R1, R2) / 4)) * 13) + (13 - Math.ceil((double)Math.max(R1, R2) / 4)));
}
}
public void map(LongWritable key, Text value, Context con) throws IOException, InterruptedException {
String line = value.toString();
String[] cards=line.split(",");
int ind_1 = GetIndex(Integer.parseInt(cards[0]),Integer.parseInt(cards[1]),Integer.parseInt(cards[0])%4 == Integer.parseInt(cards[1])%4);
int ind_2 = GetIndex(Integer.parseInt(cards[2]),Integer.parseInt(cards[3]),Integer.parseInt(cards[2])%4 == Integer.parseInt(cards[3])%4);
int ind_3 = GetIndex(Integer.parseInt(cards[4]),Integer.parseInt(cards[5]),Integer.parseInt(cards[4])%4 == Integer.parseInt(cards[4])%4);
Text outputKey = new Text(Integer.toString(ind_1) +"," + Integer.toString(ind_2) +"," + Integer.toString(ind_3));
Text outputValue = new Text(line);
con.write(outputKey, outputValue);
}
}
public static class ReduceForEquityCalculator extends Reducer<Text, IntWritable, Text, IntWritable> {
public static final int HandRankSize = 32487834;
public static int HR[] = new int[HandRankSize];
protected static final int littleEndianByteArrayToInt(byte[] b, int offset) {
return (b[offset + 3] << 24) + ((b[offset + 2] & 0xFF) << 16)
+ ((b[offset + 1] & 0xFF) << 8) + (b[offset] & 0xFF);
}
public void setup(Context context) throws IOException{
Path pt=new Path("gs://mybucket/HandRanks.dat");
Configuration conf = new Configuration();
FileSystem fs = pt.getFileSystem(conf);
int tableSize = HandRankSize*4;
byte[] b = new byte[tableSize];
try {
FileInputStream fr = new FileInputStream(fs.open(pt));
int bytesRead = fr.read(b, 0, tableSize);
if (bytesRead != tableSize) {
}
fr.close();
} catch (FileNotFoundException e) {
} catch (IOException e) {
}
for (int i = 0; i < HandRankSize; i++) {
HR[i] = littleEndianByteArrayToInt(b, i * 4);
}
}
public void reduce(Text word, Text value, Context con) throws IOException, InterruptedException {
String line = value.toString();
String[] cards_string=line.split(",");
Integer[] cards = new Integer[cards_string.length];
int i=0;
for(String str:cards_string){
cards[i]=Integer.parseInt(str);//Exception in this line
i++;
}
int c0, c1, c2, c3, c4, c5, c6;
int u0, u1, u2, u3, u4, u5;
int su0, su1, su2, su3, su4, su5;
int tu0, tu1, tu2, tu3, tu4, tu5;
u0 = HR[53+cards[0]];
u1 = HR[u0+cards[1]];
su0 = HR[53+cards[2]];
su1 = HR[su0+cards[3]];
tu0 = HR[53+cards[4]];
tu1 = HR[su0+cards[5]];
int A_B_C = 0;
int A_C_B = 0;
int A_BC = 0;
int AB_C = 0;
int AC_B = 0;
int ABC = 0;
int B_A_C = 0;
int B_C_A = 0;
int B_AC = 0;
int BC_A = 0;
int C_A_B = 0;
int C_B_A = 0;
int C_AB = 0;
for (int com1 = 1; com1 < 49; com1++) {
if (com1 != cards[0] && com1 != cards[1] && com1 != cards[2] && com1 != cards[3] && com1 != cards[4] && com1 != cards[5]) {
u2 = HR[u1+com1];
su2 = HR[su1+com1];
tu2 = HR[tu1+com1];
for (int com2 = com1 + 1; com2 < 50; com2++) {
if (com2 != cards[0] && com2 != cards[1] && com2 != cards[2] && com2 != cards[3] && com2 != cards[4] && com2 != cards[5]) {
u3 = HR[u2+com2];
su3 = HR[su2+com2];
tu3 = HR[tu2+com2];
for (int com3 = com2 + 1; com3 < 51; com3++) {
if (com3 != cards[0] && com3 != cards[1] && com3 != cards[2] && com3 != cards[3] && com3 != cards[4] && com3 != cards[5]) {
u4 = HR[u3+com3];
su4 = HR[su3+com3];
tu4 = HR[tu3+com3];
for (int com4 = com3 + 1; com4 < 52; com4++) {
if (com4 != cards[0] && com4 != cards[1] && com4 != cards[2] && com4 != cards[3] && com4 != cards[4] && com4 != cards[5]) {
u5 = HR[u4+com4];
su5 = HR[su4+com4];
tu5 = HR[tu4+com4];
for (int com5 = com4 + 1; com5 < 53; com5++) {
if (com5 != cards[0] && com5 != cards[1] && com5 != cards[2] && com5 != cards[3] && com5 != cards[4] && com5 != cards[5]) {
int rank_a = HR[u5 + com5];
int rank_b = HR[su5 + com5];
int rank_c = HR[tu5 + com5];
if (rank_a > rank_b && rank_b > rank_c) {
A_B_C++;
} else if (rank_a > rank_c && rank_c > rank_b) {
A_C_B++;
} else if (rank_a > rank_b && rank_b == rank_c) {
A_BC++;
} else if (rank_a == rank_b && rank_b > rank_c) {
AB_C++;
} else if (rank_a == rank_c && rank_c > rank_b) {
AC_B++;
} else if (rank_a == rank_c && rank_a == rank_b) {
ABC++;
} else if (rank_b > rank_a && rank_a > rank_c) {
B_A_C++;
} else if (rank_b > rank_c && rank_c > rank_a) {
B_C_A++;
} else if (rank_b > rank_a && rank_a == rank_c) {
B_AC++;
} else if (rank_b == rank_c && rank_c > rank_a) {
BC_A++;
} else if (rank_c > rank_a && rank_a > rank_b) {
C_A_B++;
} else if (rank_c > rank_b && rank_b > rank_a) {
C_B_A++;
} else {
C_AB++;
}
}
}
}
}
}
}
}
}
}
}
Text outputValue = new Text(Integer.toString(A_B_C) +"," + Integer.toString(A_C_B) + "," + Integer.toString(A_BC)+"," + Integer.toString(A_BC) +","+ Integer.toString(AB_C)+","+ Integer.toString(AC_B) +","+ Integer.toString(ABC)+"," +Integer.toString(B_A_C) +","+Integer.toString(B_C_A)+","+Integer.toString(B_AC) +","+Integer.toString(BC_A)+","+Integer.toString(C_A_B) +","+Integer.toString(C_B_A)+","+Integer.toString(C_AB) + "\n");
con.write(word, outputValue);
}
}
这将告诉JPA PK类中的@Entity
@Table(name = "rightrole")
public class RightRole extends BaseEntity<RightRolePK> {
private static final long serialVersionUID = 1L;
@EmbeddedId
protected RightRolePK rightRolePK;
@MapsId("roleID")
@JoinColumn(name = "roleID", referencedColumnName = "roleID")
@ManyToOne(fetch = FetchType.LAZY)
private Role role;
@MapsId("rightID")
@JoinColumn(name = "rightID", referencedColumnName = "rightID", insertable = false, updatable = false)
@ManyToOne(fetch = FetchType.LAZY)
private Right right;
......
}
和roleID属性中的值由关系映射控制，并且在使用主键值同步到数据库之后将设置它来自参考。您只需要确保在调用persist之前设置关系，否则它没有主键。
即使引用的对象也是复合对象，这也有效。需要引用RightRole的东西会使用RightRolePK而不是Long值：
rightID JPA还允许使用@Entity
@Table(name = "wrong_right_role")
public class WrongRightRole{
@EmbeddedId
WrongRightRoleId wrongRightRoleId;
@MapsId("rightRoleID")
@JoinColumns({
@JoinColumn(name = "roleID", referencedColumnName = "roleID"),
@JoinColumn(name = "rightID", referencedColumnName = "rightID")
})
RightRole rightRole;
}
@Embeddable
public class WrongRightRolePK implements Serializable {
private static final long serialVersionUID = 1L;
private RightRoleID rightRoleID;
.....
}
注释标记right和role映射，允许您从对象中删除@ID，并将其用作主要内容关键类而不是。
答案 1 :(得分：1)
使用@ManyToMany注释，您可以像这样定义您的实体：
作用：
@Entity
@Table(name = "role")
public class Role {
private static final long serialVersionUID = 1L;
@Id
@GeneratedValue(strategy = GenerationType.IDENTITY)
@Basic(optional = false)
@NotNull
@Column(name = "ROLE_ID", nullable = false)
private Long roleID;
@JoinTable(name = "rightrole",
joinColumns = {
@JoinColumn(name = "roleID", referencedColumnName = "ROLE_ID")},
inverseJoinColumns = {
@JoinColumn(name = "rightID", referencedColumnName = "RIGHT_ID")})
@ManyToMany
private List<Right> rightList;
}
您必须在@ManyToMany(cascade = {CascadeType.PERSIST})中写下inverseJoinColumns，否则如果删除子项，您的父数据将被删除。
右：
@Entity
@Table(name = "right")
public class Right {
private static final long serialVersionUID = 1L;
@Id
@GeneratedValue(strategy = GenerationType.IDENTITY)
@Basic(optional = false)
@NotNull
@Column(name = "RIGHT_ID",nullable = false)
private Long rightId;
}
你可以简单地说：
Role role = new Role();
List<Right> rightList = new ArrayList<>();
Right right1 = new Right();
Right right2 = new Right();
rightList.add(right1);
rightList.add(right2);
role.setRightList(rightList);
并坚持
@poinTable注释即使没有实体也会在rightrole表中插入（只要该表只有role和id的id列），这样你就可以在db中得到这样的东西了角色：
id name etc
1 roleName etc
右：
id name etc
1 rightName1 etc
2 rightName2 etc
rightrole：
roleId rightID
1 1
1 2