Enum GenericValue 

pub enum GenericValue {
    Null,
    Bool(bool),
    SmallInt(i16),
    Int(i32),
    BigInt(i64),
    Float(f32),
    Double(f64),
    Decimal(Decimal),
    String(String),
    Uuid(Uuid),
    Date(NaiveDate),
    Time(NaiveTime),
    Timestamp(NaiveDateTime),
    TimestampTz(DateTime<Utc>),
    Bytes(Vec<u8>),
    Array(Vec<GenericValue>),
    Map(BTreeMap<String, GenericValue>),
}

The canonical value vocabulary every layer trades in — the middle type between a source and a sink.

A source maps its native types into these variants; a sink maps them out to its own representation. The set is deliberately fine-grained — numerics are split by width, temporals are split into date/time/timestamp/timestamptz — so no semantic information is lost in transit: a date arrives at a sink as a Date, not an opaque string a future sink would have to guess at. Text-family Postgres types (text/varchar/citext/enum) share the one String shape because they don’t differ as values; their indexing differs, and that lives in the field’s FlussoType.

Serde is derived and format-agnostic on purpose. The derive is externally tagged ({"Date":"2024-01-01"}, {"BigInt":5}), so serialize → deserialize is a lossless identity — a Date round-trips back a Date, never collapsing to a string. That lets a queue persist or transport a value in whatever format it likes and hand it back unchanged: it goes in as a GenericValue and comes out as the same GenericValue. Core picks no format (no JSON here); the concrete encoding is the consumer’s choice.

Variants

Null

Bool(bool)

SmallInt(i16)

smallint / int2.

Int(i32)

integer / int4.

BigInt(i64)

bigint / int8.

Float(f32)

real / float4.

Double(f64)

double precision / float8.

Decimal(Decimal)

numeric / decimal — exact.

String(String)

Any text-family value (text/varchar/citext/enum).

Uuid(Uuid)

uuid.

Date(NaiveDate)

date — no time, no zone.

Time(NaiveTime)

time — no date, no zone.

Timestamp(NaiveDateTime)

timestamp — date + time, no zone.

TimestampTz(DateTime<Utc>)

timestamptz — an instant, normalized to UTC.

Bytes(Vec<u8>)

bytea.

Array(Vec<GenericValue>)

Map(BTreeMap<String, GenericValue>)

Implementations

impl GenericValue

pub fn is_bindable_scalar(&self) -> bool

Whether this value can stand as a single SQL parameter, key, or literal: true for every scalar variant, false for Null and the composite Array/Map. The one home for that rule — the Postgres source applies it when binding params, building keys, and inlining literals. Written as an exhaustive match so a new variant cannot be added without classifying it here.

Trait Implementations

impl Clone for GenericValue

fn clone(&self) -> GenericValue

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GenericValue

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for GenericValue

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for GenericValue

impl Hash for GenericValue

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 PartialEq for GenericValue

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

Tests for self and other values to be equal, and is used by ==.

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

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

impl Serialize for GenericValue

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

where __S: Serializer,

Serialize this value into the given Serde serializer.