FROM子句中的Android Room子查询无法按预期运行

时间:2019-05-20 09:00:25

标签: android-sqlite android-room

我使用https://sqliteonline.com/测试了我的查询,但无法使其与Android Room一起使用,而且我不明白为什么(相信我已经尝试过)。为什么我的DAO中的查询不起作用?另一方面,我不确定这是否是实现我要实现的目标的最佳方法,请告诉我是否应该完全改变我的方法。

而且,这是我第一次发布问题,对于我遗漏的重要事项,我深表歉意。

我用sqliteonline创建了一个测试数据库,并验证了我的查询在那儿工作。我在项目中使用的是Android Room,在build.gradle(应用程序)中包含以下几行:

// Room components
implementation 'androidx.room:room-runtime:2.1.0-beta01'
annotationProcessor 'androidx.room:room-compiler:2.1.0-beta01'
androidTestImplementation 'androidx.room:room-testing:2.1.0-beta01'

在测试环境中有效的查询:

SELECT *,
(SELECT COUNT(*) FROM groups WHERE parent_id=a.id) AS gCount,
(SELECT COUNT(*) FROM relays WHERE parent_id=a.id) AS rCount
FROM (SELECT DISTINCT * FROM worksites) a ;

我的DAO中的查询:

@Query("SELECT *, " +  
            "(SELECT COUNT(*) FROM relay_groups WHERE worksite_id=a.w_id) AS amountRelayGroups," +  
            "(SELECT COUNT(*) FROM relays_table WHERE worksite_id=a.w_id) AS amountRelays" +  
            "FROM (SELECT DISTINCT * FROM worksites_table) a")  
    LiveData<List<Worksites>> fetchAllWorksites();

我的桌子:

@Entity(tableName = "worksites_table")
    public class Worksites {

        @PrimaryKey(autoGenerate = true)
        @ColumnInfo(name = "w_id")
        private long worksite_id;

        @NonNull
        @ColumnInfo(name = "w_name")
        private String worksiteName;

        @ColumnInfo(name = "w_description")
        private String worksiteDescription;

        @Ignore
        @ColumnInfo(name = "amountRelays")
        private long amountRelays;

        @Ignore
        @ColumnInfo(name = "amountRelayGroups")
        private long amountRelayGroups;
@Entity (tableName = "relay_groups",
            foreignKeys = @ForeignKey(entity = Worksites.class,
            parentColumns = "w_id",
            childColumns = "worksite_id",
            onDelete = CASCADE))
public class RelayGroups {

    public RelayGroups(){

    }

    public RelayGroups(String name){
        this.groupName = name;
    }

    @PrimaryKey(autoGenerate = true)
    @ColumnInfo(name = "group_id")
    private long group_id;

    @NonNull
    @ColumnInfo(name = "group_name")
    private String groupName;

    @ColumnInfo(name = "description")
    private String description;

    @NonNull
    @ColumnInfo(name = "worksite_id")
    private long worksiteId;
@Entity(tableName = "relays_table",
        foreignKeys = @ForeignKey(entity = Worksites.class,
        parentColumns = "w_id",
        childColumns = "worksite_id",
        onDelete = CASCADE))
    public class Relays implements Parcelable {

    @PrimaryKey(autoGenerate = true)
    @ColumnInfo(name = "relay_id")
    private long relayId;

    @NonNull
    @ColumnInfo(name = "relay_name")
    private String relayName;

    @NonNull
    @ColumnInfo(name = "relay_number")
    private String relayNumber;

    @ColumnInfo(name = "relay_started")
    private boolean relayOnOff;

    @NonNull
    @ColumnInfo(name = "relay_type")
    private String relayType;

    @ColumnInfo(name = "registered_master")
    private boolean registeredMaster;

    @ColumnInfo(name = "registered_user")
    private boolean registeredUser;

    @ColumnInfo(name = "security_code")
    private String securityCode;

    @ColumnInfo(name = "groups_string")
    private String groupsString;

    @NonNull
    @ColumnInfo(name ="worksite_id")
    private long worksiteId;

    @ColumnInfo(name ="description")
    private String relayDescription;

我试图从表“ worksites”中获取所有列的行,并在“ relay_groups”和“ relays_table” -tables中获取匹配的worksite_ids的计数。

Java编译器打出以下错误/警告:

error: extraneous input '(' expecting {<EOF>, ';', K_ALTER, K_ANALYZE, K_ATTACH, K_BEGIN, K_COMMIT, K_CREATE, K_DELETE, K_DETACH, K_DROP, K_END,   K_EXPLAIN, K_INSERT, K_PRAGMA, K_REINDEX, K_RELEASE, K_REPLACE, K_ROLLBACK, K_SAVEPOINT, K_SELECT, K_UPDATE, K_VACUUM, K_VALUES, K_WITH, UNEXPECTED_CHAR}
no viable alternative at input 'DISTINCT * FROM worksites_table)'
The query returns some columns [amountRelayGroups, amountRelays] which are not used by PACKAGE.Worksites. You can use @ColumnInfo annotation on the  fields to specify the mapping. PACKAGE.Worksites has some fields [w_description] which are not returned by the query. If they are not  supposed to be read from the result, you can mark them with @Ignore annotation. You can suppress this warning by annotating the method with @SuppressWarnings(RoomWarnings.CURSOR_MISMATCH). Columns returned by the  query: w_name, w_id, amountRelayGroups, amountRelays. Fields in PACKAGE.Worksites: w_id, w_name, w_description.

1 个答案:

答案 0 :(得分:1)

我认为您在"FROM前缺少空格 应该是

@Query("SELECT *, " +  
            "(SELECT COUNT(*) FROM relay_groups WHERE worksite_id=a.w_id) AS amountRelayGroups," +  
            "(SELECT COUNT(*) FROM relays_table WHERE worksite_id=a.w_id) AS amountRelays" +  
            " FROM (SELECT DISTINCT * FROM worksites_table) a")