Enum mz_repr::ScalarType

pub enum ScalarType {
    Bool,
    Int16,
    Int32,
    Int64,
    UInt16,
    UInt32,
    UInt64,
    Float32,
    Float64,
    Numeric {
        max_scale: Option<NumericMaxScale>,
    },
    Date,
    Time,
    Timestamp {
        precision: Option<TimestampPrecision>,
    },
    TimestampTz {
        precision: Option<TimestampPrecision>,
    },
    Interval,
    PgLegacyChar,
    PgLegacyName,
    Bytes,
    String,
    Char {
        length: Option<CharLength>,
    },
    VarChar {
        max_length: Option<VarCharMaxLength>,
    },
    Jsonb,
    Uuid,
    Array(Box<ScalarType>),
    List {
        element_type: Box<ScalarType>,
        custom_id: Option<GlobalId>,
    },
    Record {
        fields: Vec<(ColumnName, ColumnType)>,
        custom_id: Option<GlobalId>,
    },
    Oid,
    Map {
        value_type: Box<ScalarType>,
        custom_id: Option<GlobalId>,
    },
    RegProc,
    RegType,
    RegClass,
    Int2Vector,
    MzTimestamp,
    Range {
        element_type: Box<ScalarType>,
    },
    MzAclItem,
    AclItem,
}

The type of a Datum.

There is a direct correspondence between Datum variants and ScalarType variants.

Variants

Bool

The type of Datum::True and Datum::False.

Int16

The type of Datum::Int16.

Int32

The type of Datum::Int32.

Int64

The type of Datum::Int64.

UInt16

The type of Datum::UInt16.

UInt32

The type of Datum::UInt32.

UInt64

The type of Datum::UInt64.

Float32

The type of Datum::Float32.

Float64

The type of Datum::Float64.

Numeric

Fields

max_scale: Option<NumericMaxScale>

The type of Datum::Numeric.

Numeric values cannot exceed NUMERIC_DATUM_MAX_PRECISION digits of precision.

This type additionally specifies the maximum scale of the decimal. The scale specifies the number of digits after the decimal point.

Date

The type of Datum::Date.

Time

The type of Datum::Time.

Timestamp

Fields

precision: Option<TimestampPrecision>

The type of Datum::Timestamp.

TimestampTz

Fields

precision: Option<TimestampPrecision>

The type of Datum::TimestampTz.

Interval

The type of Datum::Interval.

PgLegacyChar

A single byte character type backed by a Datum::UInt8.

PostgreSQL calls this type "char". Note the quotes, which distinguish it from the type ScalarType::Char.

PgLegacyName

A character type for storing identifiers of no more than 64 characters in length.

PostgreSQL uses this type to represent the names of objects in the system catalog.

Bytes

The type of Datum::Bytes.

String

The type of Datum::String.

Char

Fields

length: Option<CharLength>

Stored as Datum::String, but expresses a fixed-width, blank-padded string.

Note that a length of None is used in special cases, such as creating lists.

VarChar

Fields

max_length: Option<VarCharMaxLength>

Stored as Datum::String, but can optionally express a limit on the string’s length.

Jsonb

The type of a datum that may represent any valid JSON value.

Valid datum variants for this type are:

• Datum::JsonNull
• Datum::False
• Datum::True
• Datum::String
• Datum::Numeric
• Datum::List
• Datum::Map

Uuid

The type of Datum::Uuid.

Array(Box<ScalarType>)

The type of Datum::Array.

Elements within the array are of the specified type. It is illegal for the element type to be itself an array type. Array elements may always be Datum::Null.

List

Fields

element_type: Box<ScalarType>

custom_id: Option<GlobalId>

The type of Datum::List.

Elements within the list are of the specified type. List elements may always be Datum::Null.

Record

Fields

fields: Vec<(ColumnName, ColumnType)>

The names and types of the fields of the record, in order from left to right.

custom_id: Option<GlobalId>

An ordered and named sequence of datums.

Oid

A PostgreSQL object identifier.

Map

Fields

value_type: Box<ScalarType>

custom_id: Option<GlobalId>

The type of Datum::Map

Keys within the map are always of type ScalarType::String. Values within the map are of the specified type. Values may always be Datum::Null.

RegProc

A PostgreSQL function name.

RegType

A PostgreSQL type name.

RegClass

A PostgreSQL class name.

Int2Vector

A vector on small ints; this is a legacy type in PG used primarily in the catalog.

MzTimestamp

A Materialize timestamp.

Range

Fields

element_type: Box<ScalarType>

MzAclItem

The type of Datum::MzAclItem

AclItem

The type of Datum::AclItem

Implementations

impl<'a> ScalarType

pub fn unwrap_numeric_max_scale(&self) -> Option<NumericMaxScale>

Returns the contained numeric maximum scale.

Panics

Panics if the scalar type is not ScalarType::Numeric.

pub fn unwrap_timestamp_precision(&self) -> Option<TimestampPrecision>

Returns the contained timestamp precision.

Panics

Panics if the scalar type is not ScalarType::Timestamp or ScalarType::TimestampTz.

pub fn unwrap_list_element_type(&self) -> &ScalarType

Returns the ScalarType of elements in a ScalarType::List.

Panics

Panics if called on anything other than a ScalarType::List.

pub fn unwrap_list_nth_layer_type(&self, layer: usize) -> &ScalarType

Returns the ScalarType of elements in the nth layer a ScalarType::List.

For example, in an int list list, the:

• 0th layer is int list list
• 1st layer is int list
• 2nd layer is int

Panics

Panics if the nth-1 layer is anything other than a ScalarType::List.

pub fn unwrap_record_element_type(&self) -> Vec<&ScalarType>

Returns vector of ScalarType elements in a ScalarType::Record.

Panics

Panics if called on anything other than a ScalarType::Record.

pub fn unwrap_list_n_layers(&self) -> usize

Returns number of dimensions/axes (also known as “rank”) on a ScalarType::List.

Panics

Panics if called on anything other than a ScalarType::List.

pub fn without_modifiers(&self) -> ScalarType

Returns self with any type modifiers removed.

Namely, this should set optional scales or limits to None.

pub fn unwrap_array_element_type(&self) -> &ScalarType

Returns the ScalarType of elements in a ScalarType::Array or the elements of a vector type, e.g. ScalarType::Int16 for ScalarType::Int2Vector.

Panics

Panics if called on anything other than a ScalarType::Array or ScalarType::Int2Vector.

pub fn unwrap_collection_element_type(&self) -> &ScalarType

Returns the ScalarType of elements in a ScalarType::Array, ScalarType::Int2Vector, or ScalarType::List.

Panics

Panics if called on anything other than a ScalarType::Array, ScalarType::Int2Vector, or ScalarType::List.

pub fn unwrap_map_value_type(&self) -> &ScalarType

Returns the ScalarType of values in a ScalarType::Map.

Panics

Panics if called on anything other than a ScalarType::Map.

pub fn unwrap_char_length(&self) -> Option<CharLength>

Returns the length of a ScalarType::Char.

Panics

Panics if called on anything other than a ScalarType::Char.

pub fn unwrap_varchar_max_length(&self) -> Option<VarCharMaxLength>

Returns the max length of a ScalarType::VarChar.

Panics

Panics if called on anything other than a ScalarType::VarChar.

pub fn unwrap_range_element_type(&self) -> &ScalarType

Returns the ScalarType of elements in a ScalarType::Range.

Panics

Panics if called on anything other than a ScalarType::Map.

pub fn near_match(&self) -> Option<&'static ScalarType>

Returns a “near match” of self, which are types that are implicitly castable from self and offer a means to leverage Materialize’s type system to achieve more reasonable approaches to unifying types.

However, it’s very important to not blithely accept the near_match, which can be suboptimal/unnecessary, e.g. in the case of an already homogeneous group.

The feature is preferrable in MZ, but unnecessary in PG because PG’s type system offers totally linear progression through the complexity of types. e.g. with numbers, there is a linear progression in the domain each can represent. However, MZ’s support for unsigned integers create a non-linear type system, i.e. while the magnitude of Int32 and UInt32’s domains are the same, they are not equal.

Without this feature, Materialize will:

• Guess that a mixute of the same width of int and uint cannot be coerced to a homogeneous type.
• Select the Float64 based version of common binary functions (e.g. =), which introduces an unexpected float cast to integer values.

Note that if adding any near matches besides unsigned ints, consider extending/generalizing how guess_best_common_type uses this function.

pub const fn nullable(self, nullable: bool) -> ColumnType

Derives a column type from this scalar type with the specified nullability.

pub fn is_vec(&self) -> bool

Returns whether or not self is a vector-like type, i.e. ScalarType::Array, ScalarType::Int2Vector, or ScalarType::List, irrespective of its element type.

pub fn is_custom_type(&self) -> bool

pub fn base_eq(&self, other: &ScalarType) -> bool

Determines equality among scalar types that acknowledges custom OIDs, but ignores other embedded values.

In most situations, you want to use base_eq rather than ScalarType’s implementation of Eq. base_eq expresses the semantics of direct type interoperability whereas Eq expresses an exact comparison between the values.

For instance, base_eq signals that e.g. two ScalarType::Numeric values can be added together, irrespective of their embedded scale. In contrast, two Numeric values with different scales are never Eq to one another.

pub fn structural_eq(&self, other: &ScalarType) -> bool

pub fn eq_inner(&self, other: &ScalarType, structure_only: bool) -> bool

pub fn interesting_datums(&self) -> impl Iterator<Item = Datum<'static>>

Returns various interesting datums for a ScalarType (max, min, 0 values, etc.).

pub fn enumerate() -> &'static [Self]

Returns all non-parameterized types and some versions of some parameterized types.

pub fn array_of_self_elem_type(self) -> Result<ScalarType, ScalarType>

Returns the appropriate element type for making a ScalarType::Array whose elements are of self.

If the type is not compatible with making an array, returns in the error position.

Trait Implementations

impl Arbitrary for ScalarType

type Parameters = ()

The type of parameters that arbitrary_with accepts for configuration of the generated Strategy. Parameters must implement Default.

type Strategy = BoxedStrategy<ScalarType>

The type of Strategy used to generate values of type Self.

fn arbitrary_with(_: Self::Parameters) -> Self::Strategy

Generates a Strategy for producing arbitrary values of type the implementing type (Self). The strategy is passed the arguments given in args.

fn arbitrary() -> Self::Strategy

Generates a Strategy for producing arbitrary values of type the implementing type (Self).

impl Clone for ScalarType

fn clone(&self) -> ScalarType

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for ScalarType

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ScalarType

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl<'__enum_kinds1> From<&'__enum_kinds1 ScalarType> for ScalarBaseType

fn from(_value: &'__enum_kinds1 ScalarType) -> Self

Converts to this type from the input type.

impl<'__enum_kinds1> From<ScalarType> for ScalarBaseType

fn from(value: ScalarType) -> Self

Converts to this type from the input type.

impl Hash for ScalarType

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl MzReflect for ScalarType

fn add_to_reflected_type_info(rti: &mut ReflectedTypeInfo)

Adds names and types of the fields of the struct or enum to rti.

impl Ord for ScalarType

fn cmp(&self, other: &ScalarType) -> Ordering

This method returns an Ordering between self and other.

fn max(self, other: Self) -> Selfwhere Self: Sized,

Compares and returns the maximum of two values.

fn min(self, other: Self) -> Selfwhere Self: Sized,

Compares and returns the minimum of two values.

fn clamp(self, min: Self, max: Self) -> Selfwhere Self: Sized + PartialOrd<Self>,

Restrict a value to a certain interval.

impl PartialEq<ScalarType> for ScalarType

fn eq(&self, other: &ScalarType) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl PartialOrd<ScalarType> for ScalarType

fn partial_cmp(&self, other: &ScalarType) -> Option<Ordering>

This method returns an ordering between self and other values if one exists.

fn lt(&self, other: &Rhs) -> bool

This method tests less than (for self and other) and is used by the < operator.

fn le(&self, other: &Rhs) -> bool

This method tests less than or equal to (for self and other) and is used by the <= operator.

fn gt(&self, other: &Rhs) -> bool

This method tests greater than (for self and other) and is used by the > operator.

fn ge(&self, other: &Rhs) -> bool

This method tests greater than or equal to (for self and other) and is used by the >= operator.

impl RustType<ProtoScalarType> for ScalarType

fn into_proto(&self) -> ProtoScalarType

Convert a Self into a Proto value.

fn from_proto(proto: ProtoScalarType) -> Result<Self, TryFromProtoError>

Consume and convert a Proto back into a Self value.

impl Serialize for ScalarType

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Eq for ScalarType

impl StructuralEq for ScalarType

impl StructuralPartialEq for ScalarType