如何在房间数据库中查询和过滤关系实体
How to query and filter relation entity in room database
例如：
用户：
@Entity(tableName = "user")
data class UserEntity(
@PrimaryKey
@ColumnInfo(name = "id") val id: String,
@ColumnInfo(name = "username") val username: String,
@ColumnInfo(name = "name") val name: String,
Post:
@Entity(
tableName = "post",
foreignKeys = [
ForeignKey(
entity = UserEntity::class,
parentColumns = ["id"],
childColumns = ["user_id"],
onDelete = ForeignKey.CASCADE,
onUpdate = ForeignKey.CASCADE
)
],
indices = [Index(value = ["user_id"])]
)
data class PostEntity(
@PrimaryKey
@ColumnInfo(name = "id") var id: String,
@ColumnInfo(name = "user_id") var userId: String,
@ColumnInfo(name = "body") val body: String,
@ColumnInfo(name = "like") val like: Int,
@ColumnType(name = "type") val type: String,
)
数据
data class Data(
@Embedded
val user: UserEntity,
@Relation(parentColumn = "id", entityColumn = "user_id")
val post: List<PostEntity> = emptyList(),
)
如果我使用 SELECT * FROM user 我得到了想要的数据（一个用户和所有 post），但是我如何过滤特定类型的 post，比如 WHERE post.type = 'sth' 这可能吗？
but how can i filter the post for a specific type,
这完全取决于您要过滤的内容。您可能需要匹配过滤器但具有所有 post 的数据对象（与类型无关），在这种情况下您可以使用：-
@Transaction
@Query("SELECT * FROM user JOIN post ON user.id = user_id WHERE post.type = :type")
abstract fun getAllDataFiltered(type: String): List<Data>
- 你会在什么地方使用
var mylist = yourdao.getAllDataFiltered("sth")
但是，由于 post 和用户的 id 列都被命名为 id 那么歧义会干扰（用户id 成为 post id，因此没有底层 post 对象被提取。
如果您将 PostEntity 更改为：-
@Entity(
tableName = "post",
foreignKeys = [
ForeignKey(
entity = UserEntity::class,
parentColumns = ["id"],
childColumns = ["user_id"],
onDelete = ForeignKey.CASCADE,
onUpdate = ForeignKey.CASCADE
)
],
indices = [Index(value = ["user_id"])]
)
data class PostEntity(
@PrimaryKey
@ColumnInfo(name = "postid") var id: String, //<<<<<<<<<< CHANGED
@ColumnInfo(name = "user_id") var userId: String,
@ColumnInfo(name = "body") val body: String,
@ColumnInfo(name = "like") val like: Int,
@ColumnInfo(name = "type") val type: String
)
然后消除了歧义，数据对象 returned 包括所有具有 post 类型 sth 的相应用户的 posts .
如果您只想 return 编辑只有过滤后的 post 的数据对象，那么您必须绕过 Room 的 returning FULL/COMPLETE 技术相关对象。
如果您将 @Dao class 设为抽象 class 而不是接口，那么您将使用 @Query 例如 :-
@Query("SELECT * FROM post WHERE user_id=:userid AND type=:type")
abstract fun getPostsPerUserFiltered(userid: String, type: String): List<PostEntity>
连同函数，例如：-
fun getFullyFiltered(type: String): List<Data> {
var rv: ArrayList<Data> = arrayListOf()
for(d: Data in getAllDataFiltered(type)) {
rv.add(Data(d.user,post = getPostsPerUserFiltered(d.user.id,type)))
}
return rv
}
这会过滤数据对象 return，但会丢弃整个 PostEntity 列表（即每个 post，与过滤器无关），然后应用过滤后的 post。
如果您想要所有用户，但只有 post 匹配（因此可能是 post 的空列表），那么您可以使用如下函数：-
fun getAllFilteredData(type: String): List<Data> {
var rv: ArrayList<Data> = arrayListOf()
for(u: UserEntity in getUsers()) {
rv.add(Data(user = u, getPostsPerUserFiltered(u.id,type)))
}
return rv
}
即不对用户应用过滤，只对 posts.
然后使用上面的内容（注意使用更改的列名（postid 而不是 id））然后考虑以下内容（使用了非常标准的@Database class） :-
class MainActivity : AppCompatActivity() {
lateinit var db: TheDatabase
lateinit var dao: AllDao
val TAG: String = "DBINFO"
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(R.layout.activity_main)
db = TheDatabase.getInstance(this)
dao = db.getAllDao()
dao.insert(UserEntity(id = "User1","User001","Mary"))
dao.insert(UserEntity(id = "User2",username = "User002",name = "Sue"))
dao.insert(UserEntity(id = "User3",username = "User003",name = "Tom"))
dao.insert(PostEntity(id = "post1",userId = "User1",body = "post1 blah",type = "xxx",like = 0))
dao.insert(PostEntity(id ="post2", userId = "User2",body ="post2 blah", type = "sth",like = 1))
dao.insert(PostEntity(id = "post3", userId = "User1", body = "post3 blah", type = "sth", like = 3))
/*
No filtering applied
*/
for(d: Data in dao.getAllData()) {
logData(d,"ALL")
}
/*
Return FULL Data objects (i.e. with ALL posts) but only those that
have a post or posts that match the filter
*/
for (d: Data in dao.getAllDataFiltered("sth")) {
logData(d,"JOIN")
}
/*
Return partial Data objects, but only for those that
have a post that matches the type
*/
for (d: Data in dao.getFullyFiltered("sth")) {
logData(d,"FULL")
}
/*
Return all partial Data Objects but with partial posts.
*/
for (d: Data in dao.getAllFilteredData("sth")) {
logData(d,"POST")
}
}
private fun logData(data: Data,tagSuffix: String) {
Log.d(TAG + tagSuffix,"Data for user ${data.user.id}, Name is ${data.user.name} etc")
for (p: PostEntity in data.post) {
Log.d(TAG + tagSuffix,"\t Post is ${p.id} Type is ${p.type} body is:-\n\t\t${p.body}")
}
}
}
输出到日志的结果是：-
完全没有过滤：-
2021-09-17 11:22:00.722 D/DBINFOALL: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.722 D/DBINFOALL: Post is post1 Type is xxx body is:-
post1 blah
2021-09-17 11:22:00.723 D/DBINFOALL: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:22:00.723 D/DBINFOALL: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.723 D/DBINFOALL: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.723 D/DBINFOALL: Data for user User3, Name is Tom etc
只有用户被过滤，因为 Room 获得用户的所有 post
:-
2021-09-17 11:22:00.728 D/DBINFOJOIN: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.728 D/DBINFOJOIN: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.728 D/DBINFOJOIN: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.728 D/DBINFOJOIN: Post is post1 Type is xxx body is:-
post1 blah
2021-09-17 11:22:00.728 D/DBINFOJOIN: Post is post3 Type is sth body is:-
post3 blah
完全过滤：-
2021-09-17 11:22:00.738 D/DBINFOFULL: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.738 D/DBINFOFULL: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.738 D/DBINFOFULL: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.738 D/DBINFOFULL: Post is post3 Type is sth body is:-
post3 blah
所有用户，但过滤 posts
2021-09-17 11:22:00.744 D/DBINFOPOST: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.744 D/DBINFOPOST: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:22:00.744 D/DBINFOPOST: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.744 D/DBINFOPOST: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.744 D/DBINFOPOST: Data for user User3, Name is Tom etc
如果 PostEntity 列改回 id，则结果为
:-
2021-09-17 11:27:00.661 D/DBINFOALL: Data for user User1, Name is Mary etc
2021-09-17 11:27:00.661 D/DBINFOALL: Post is post1 Type is xxx body is:-
post1 blah
2021-09-17 11:27:00.661 D/DBINFOALL: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:27:00.661 D/DBINFOALL: Data for user User2, Name is Sue etc
2021-09-17 11:27:00.662 D/DBINFOALL: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:27:00.662 D/DBINFOALL: Data for user User3, Name is Tom etc
2021-09-17 11:27:00.664 D/DBINFOJOIN: Data for user post2, Name is Sue etc
2021-09-17 11:27:00.664 D/DBINFOJOIN: Data for user post3, Name is Mary etc
2021-09-17 11:27:00.672 D/DBINFOFULL: Data for user post2, Name is Sue etc
2021-09-17 11:27:00.672 D/DBINFOFULL: Data for user post3, Name is Mary etc
2021-09-17 11:27:00.676 D/DBINFOPOST: Data for user User1, Name is Mary etc
2021-09-17 11:27:00.676 D/DBINFOPOST: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:27:00.676 D/DBINFOPOST: Data for user User2, Name is Sue etc
2021-09-17 11:27:00.676 D/DBINFOPOST: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:27:00.677 D/DBINFOPOST: Data for user User3, Name is Tom etc
- 即注意用户是post的ID，因此没有基础post。
- 你可以使用
@Embedded(prefix = "a_suitable_prefix")。但是，您随后必须使用 AS 更改查询中用户的列名（前缀为 table），使用明确的列名要简单得多。
- 第 4 个，returning 所有用户但仅过滤 posts 不受影响，因为它不使用数据 POJO，这是导致用户 ID 出现歧义的地方post ID。
例如：
用户：
@Entity(tableName = "user")
data class UserEntity(
@PrimaryKey
@ColumnInfo(name = "id") val id: String,
@ColumnInfo(name = "username") val username: String,
@ColumnInfo(name = "name") val name: String,
Post:
@Entity(
tableName = "post",
foreignKeys = [
ForeignKey(
entity = UserEntity::class,
parentColumns = ["id"],
childColumns = ["user_id"],
onDelete = ForeignKey.CASCADE,
onUpdate = ForeignKey.CASCADE
)
],
indices = [Index(value = ["user_id"])]
)
data class PostEntity(
@PrimaryKey
@ColumnInfo(name = "id") var id: String,
@ColumnInfo(name = "user_id") var userId: String,
@ColumnInfo(name = "body") val body: String,
@ColumnInfo(name = "like") val like: Int,
@ColumnType(name = "type") val type: String,
)
数据
data class Data(
@Embedded
val user: UserEntity,
@Relation(parentColumn = "id", entityColumn = "user_id")
val post: List<PostEntity> = emptyList(),
)
如果我使用 SELECT * FROM user 我得到了想要的数据（一个用户和所有 post），但是我如何过滤特定类型的 post，比如 WHERE post.type = 'sth' 这可能吗？
but how can i filter the post for a specific type,
这完全取决于您要过滤的内容。您可能需要匹配过滤器但具有所有 post 的数据对象（与类型无关），在这种情况下您可以使用：-
@Transaction
@Query("SELECT * FROM user JOIN post ON user.id = user_id WHERE post.type = :type")
abstract fun getAllDataFiltered(type: String): List<Data>
- 你会在什么地方使用
var mylist = yourdao.getAllDataFiltered("sth")
但是，由于 post 和用户的 id 列都被命名为 id 那么歧义会干扰（用户id 成为 post id，因此没有底层 post 对象被提取。
如果您将 PostEntity 更改为：-
@Entity(
tableName = "post",
foreignKeys = [
ForeignKey(
entity = UserEntity::class,
parentColumns = ["id"],
childColumns = ["user_id"],
onDelete = ForeignKey.CASCADE,
onUpdate = ForeignKey.CASCADE
)
],
indices = [Index(value = ["user_id"])]
)
data class PostEntity(
@PrimaryKey
@ColumnInfo(name = "postid") var id: String, //<<<<<<<<<< CHANGED
@ColumnInfo(name = "user_id") var userId: String,
@ColumnInfo(name = "body") val body: String,
@ColumnInfo(name = "like") val like: Int,
@ColumnInfo(name = "type") val type: String
)
然后消除了歧义，数据对象 returned 包括所有具有 post 类型 sth 的相应用户的 posts .
如果您只想 return 编辑只有过滤后的 post 的数据对象，那么您必须绕过 Room 的 returning FULL/COMPLETE 技术相关对象。
如果您将 @Dao class 设为抽象 class 而不是接口，那么您将使用 @Query 例如 :-
@Query("SELECT * FROM post WHERE user_id=:userid AND type=:type")
abstract fun getPostsPerUserFiltered(userid: String, type: String): List<PostEntity>
连同函数，例如：-
fun getFullyFiltered(type: String): List<Data> {
var rv: ArrayList<Data> = arrayListOf()
for(d: Data in getAllDataFiltered(type)) {
rv.add(Data(d.user,post = getPostsPerUserFiltered(d.user.id,type)))
}
return rv
}
这会过滤数据对象 return，但会丢弃整个 PostEntity 列表（即每个 post，与过滤器无关），然后应用过滤后的 post。
如果您想要所有用户，但只有 post 匹配（因此可能是 post 的空列表），那么您可以使用如下函数：-
fun getAllFilteredData(type: String): List<Data> {
var rv: ArrayList<Data> = arrayListOf()
for(u: UserEntity in getUsers()) {
rv.add(Data(user = u, getPostsPerUserFiltered(u.id,type)))
}
return rv
}
即不对用户应用过滤，只对 posts.
然后使用上面的内容（注意使用更改的列名（postid 而不是 id））然后考虑以下内容（使用了非常标准的@Database class） :-
class MainActivity : AppCompatActivity() {
lateinit var db: TheDatabase
lateinit var dao: AllDao
val TAG: String = "DBINFO"
override fun onCreate(savedInstanceState: Bundle?) {
super.onCreate(savedInstanceState)
setContentView(R.layout.activity_main)
db = TheDatabase.getInstance(this)
dao = db.getAllDao()
dao.insert(UserEntity(id = "User1","User001","Mary"))
dao.insert(UserEntity(id = "User2",username = "User002",name = "Sue"))
dao.insert(UserEntity(id = "User3",username = "User003",name = "Tom"))
dao.insert(PostEntity(id = "post1",userId = "User1",body = "post1 blah",type = "xxx",like = 0))
dao.insert(PostEntity(id ="post2", userId = "User2",body ="post2 blah", type = "sth",like = 1))
dao.insert(PostEntity(id = "post3", userId = "User1", body = "post3 blah", type = "sth", like = 3))
/*
No filtering applied
*/
for(d: Data in dao.getAllData()) {
logData(d,"ALL")
}
/*
Return FULL Data objects (i.e. with ALL posts) but only those that
have a post or posts that match the filter
*/
for (d: Data in dao.getAllDataFiltered("sth")) {
logData(d,"JOIN")
}
/*
Return partial Data objects, but only for those that
have a post that matches the type
*/
for (d: Data in dao.getFullyFiltered("sth")) {
logData(d,"FULL")
}
/*
Return all partial Data Objects but with partial posts.
*/
for (d: Data in dao.getAllFilteredData("sth")) {
logData(d,"POST")
}
}
private fun logData(data: Data,tagSuffix: String) {
Log.d(TAG + tagSuffix,"Data for user ${data.user.id}, Name is ${data.user.name} etc")
for (p: PostEntity in data.post) {
Log.d(TAG + tagSuffix,"\t Post is ${p.id} Type is ${p.type} body is:-\n\t\t${p.body}")
}
}
}
输出到日志的结果是：-
完全没有过滤：-
2021-09-17 11:22:00.722 D/DBINFOALL: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.722 D/DBINFOALL: Post is post1 Type is xxx body is:-
post1 blah
2021-09-17 11:22:00.723 D/DBINFOALL: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:22:00.723 D/DBINFOALL: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.723 D/DBINFOALL: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.723 D/DBINFOALL: Data for user User3, Name is Tom etc
只有用户被过滤，因为 Room 获得用户的所有 post :-
2021-09-17 11:22:00.728 D/DBINFOJOIN: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.728 D/DBINFOJOIN: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.728 D/DBINFOJOIN: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.728 D/DBINFOJOIN: Post is post1 Type is xxx body is:-
post1 blah
2021-09-17 11:22:00.728 D/DBINFOJOIN: Post is post3 Type is sth body is:-
post3 blah
完全过滤：-
2021-09-17 11:22:00.738 D/DBINFOFULL: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.738 D/DBINFOFULL: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.738 D/DBINFOFULL: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.738 D/DBINFOFULL: Post is post3 Type is sth body is:-
post3 blah
所有用户，但过滤 posts
2021-09-17 11:22:00.744 D/DBINFOPOST: Data for user User1, Name is Mary etc
2021-09-17 11:22:00.744 D/DBINFOPOST: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:22:00.744 D/DBINFOPOST: Data for user User2, Name is Sue etc
2021-09-17 11:22:00.744 D/DBINFOPOST: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:22:00.744 D/DBINFOPOST: Data for user User3, Name is Tom etc
如果 PostEntity 列改回 id，则结果为
:-
2021-09-17 11:27:00.661 D/DBINFOALL: Data for user User1, Name is Mary etc
2021-09-17 11:27:00.661 D/DBINFOALL: Post is post1 Type is xxx body is:-
post1 blah
2021-09-17 11:27:00.661 D/DBINFOALL: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:27:00.661 D/DBINFOALL: Data for user User2, Name is Sue etc
2021-09-17 11:27:00.662 D/DBINFOALL: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:27:00.662 D/DBINFOALL: Data for user User3, Name is Tom etc
2021-09-17 11:27:00.664 D/DBINFOJOIN: Data for user post2, Name is Sue etc
2021-09-17 11:27:00.664 D/DBINFOJOIN: Data for user post3, Name is Mary etc
2021-09-17 11:27:00.672 D/DBINFOFULL: Data for user post2, Name is Sue etc
2021-09-17 11:27:00.672 D/DBINFOFULL: Data for user post3, Name is Mary etc
2021-09-17 11:27:00.676 D/DBINFOPOST: Data for user User1, Name is Mary etc
2021-09-17 11:27:00.676 D/DBINFOPOST: Post is post3 Type is sth body is:-
post3 blah
2021-09-17 11:27:00.676 D/DBINFOPOST: Data for user User2, Name is Sue etc
2021-09-17 11:27:00.676 D/DBINFOPOST: Post is post2 Type is sth body is:-
post2 blah
2021-09-17 11:27:00.677 D/DBINFOPOST: Data for user User3, Name is Tom etc
- 即注意用户是post的ID，因此没有基础post。
- 你可以使用
@Embedded(prefix = "a_suitable_prefix")。但是，您随后必须使用AS更改查询中用户的列名（前缀为 table），使用明确的列名要简单得多。 - 第 4 个，returning 所有用户但仅过滤 posts 不受影响，因为它不使用数据 POJO，这是导致用户 ID 出现歧义的地方post ID。