Major Description
Changing query more intuitively by using `columnName` instead of `from/to` in label field name
When fetching a Label Associated with a ServiceColumn, `from/to` Is used as the Label field.
This indirectly represents the `source/target Column` in the Label.
You can also query the next step based on the corresponding `from/to`.
ColumnName defined in the Label schema to create a more intuitive query than using `from/to`.
For example, suppose you have the following Services, ServiceColumns, and Label.
service = Service(name: kakao) userColumn = ServiceColumn(serviceName: kakao, name: user) movieColumn = ServiceColumn(serviceName: kakao, name: movie) Label(name: like, sourceServiceColumn: userColumn, targetServiceColumn: movieColumn)
Let's take an example of a query that comes from a user who likes movies that user 'daewon' likes.
As is
query {
kakao {
user(id: "daewon") {
like(direction: out) { # By Label(like) shcmea, direction is fixed in `out`
score
to { # `to` Represents the targetServiceColumn (movie) of `like` label.
id
like(direction: in) { # By Label(like) shcmea, direction is fixed in `in`
score
from { # In the case of the in direction, it is confused because the directions `from` and `to` are relatively determined.
id
age
}
}
}
}
}
}
}
To be
query {
kakao {
user(id: "daewon") {
like { # For a Label(like) started with serviceColumn(user), the direction is fixed to `out`.
score
movie {
id
like { # For a Label(like) started with serviceColumn(movie), the direction is fixed to `in`.
score
user {
id
age
}
}
}
}
}
}
}
As you can see in the example above, we can know in advance how to query `like` label according to the starting column(user/movie), and also use field name instead of `from/to` to use columnName.
Attachments
Issue Links
- links to
