Struct QueryField 

#[non_exhaustive]
pub struct QueryField {

    pub name: String,
    pub table_name: String,
    pub schema_name: String,
    pub type_name: String,
    pub local_type_name: String,
    pub return_type: String,
    pub is_list: bool,
    pub is_by_pk: bool,
    pub pk_columns: Vec<(String, String)>,
    pub description: Option<String>,
    pub aggregate_name: Option<String>,
    pub aggregate_description: Option<String>,
    pub aggregates: bool,
    pub rows: bool,
}

A query field for a table (e.g., users, userByPk).

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

name: String

Field name (e.g., “users”)

table_name: String

Table name

schema_name: String

Schema the table lives in

type_name: String

GraphQL object type name (e.g., “Users”)

local_type_name: String

Base name for helper types owned by this table.

return_type: String

GraphQL return type

is_list: bool

Whether this returns a list

is_by_pk: bool

Whether this is a “by PK” query

pk_columns: Vec<(String, String)>

Primary key columns, as (column name, PostgreSQL type).

Populated for by-PK queries so the resolver can filter on the table’s actual key rather than assuming a column called id. Empty for list queries.

description: Option<String>

Field description

aggregate_name: Option<String>

The name given to this table’s aggregate root, if one was.

The aggregate field is built beside the list field rather than being a QueryField of its own, so the name it was given rides along with the list field that spawns it.

aggregate_description: Option<String>

The description given to that aggregate root, if one was. Some("") means metadata said it has none.

aggregates: bool

Whether the aggregate root is built at all.

The one thing a role’s view of the schema cache cannot say. Dropping a table or a column is a fact about the database as this role sees it; whether that role may count the rows it can already read is a fact about the permission, and Hasura keeps the two apart because counting rows you cannot see is a way of seeing them. True unless a select permission says otherwise, which is what leaves an unconfigured server exactly as it was.

rows: bool

Whether the rows themselves can be asked for.

False for a table a role may count and not read: Hasura’s way of granting “how many” without granting “which” is a select permission naming no columns, and such a table has an aggregate root and no row type to hang a list root on. The field is still carried so that the aggregate root beside it is built.

Implementations

impl QueryField

pub fn list(table: &Table) -> Self

Create a list query field (e.g., users), named after the table.

pub fn list_named(table: &Table, base_name: &str) -> Self

Create a list query field using an explicit base name.

pub fn list_typed( table: &Table, root_base_name: &str, type_name: &str, local_type_name: &str, ) -> Self

Create a list query field with separate root, row type and helper names.

pub fn by_pk(table: &Table) -> Option<Self>

Create a by-PK query field (e.g., userByPk), named after the table.

pub fn by_pk_named(table: &Table, base_name: &str) -> Option<Self>

Create a by-PK query field using an explicit base name.

pub fn by_pk_typed( table: &Table, root_base_name: &str, type_name: &str, local_type_name: &str, ) -> Option<Self>

Create a by-PK query field with separate root, row type and helper names.

Trait Implementations

impl Clone for QueryField

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for QueryField

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.