Struct SqliteDialect 

pub struct SqliteDialect;

Trait Implementations

impl Dialect for SqliteDialect

fn name(&self) -> &'static str

Short name for log messages and errors (e.g. “postgres”, “mysql”, “sqlite”).

fn ddl_type(&self, t: &CanonicalType) -> String

DDL type string for CREATE TABLE (e.g. “TIMESTAMPTZ”, “DATETIME”, “TEXT”).

fn cast_name(&self, _t: &CanonicalType) -> Option<String>

Type name used in parameter cast expressions, or None when no cast is needed. Postgres: becomes $n::cast. MySQL/SQLite: cast is omitted (binding handles type).

fn type_category(&self, t: &CanonicalType) -> TypeCategory

Broad category for RSQL operator validation.

fn type_support(&self, t: &CanonicalType) -> TypeSupport

How well this dialect supports the canonical type.

fn quote_ident(&self, s: &str) -> String

Wrap an identifier in dialect-specific delimiters. Postgres/SQLite: double-quotes. MySQL: backticks.

fn placeholder(&self, _n: usize) -> String

Positional placeholder for the n-th parameter (1-based). Postgres: $1. MySQL/SQLite: ?.

fn cast_expr(&self, placeholder: &str, _cast: &str) -> String

Wrap a placeholder with a type cast where required. Postgres: $1::uuid. MySQL/SQLite: placeholder returned unchanged.

fn now_fn(&self) -> &'static str

Current-timestamp function name/expression.

fn uuid_default_expr(&self) -> &'static str

Expression that generates a random UUID as a column DEFAULT.

fn returning_clause(&self, cols: &str) -> String

RETURNING clause appended to INSERT/UPDATE/DELETE, or empty string when unsupported.

fn upsert_conflict(&self, conflict_cols: &[&str], set_pairs: &str) -> String

Upsert conflict suffix. conflict_cols: columns that identify the conflict. set_pairs: pre-built “col = value” pairs for the update branch.

fn to_one_subquery(&self, col_exprs: &[String], from_clause: &str) -> String

Build a scalar subquery returning a single JSON object for a to-one include. col_exprs: already-quoted column expressions. from_clause: "schema"."table" WHERE ... fragment.

fn to_many_subquery(&self, col_exprs: &[String], from_clause: &str) -> String

Build a scalar subquery returning a JSON array for a to-many include.

fn json_extract_text(&self, col: &str, key: &str) -> String

Extract a top-level key from a JSON/JSONB column as text. col is an already-quoted column expression; key is the raw JSON key (escaped here). Postgres: (col ->> 'key'). MySQL/SQLite: col->>'$.key'.

fn json_extract_typed(&self, col: &str, key: &str, t: &CanonicalType) -> String

Extract a JSON key and cast it to t so comparisons and ORDER BY are type-correct. For text-like types this is equivalent to Dialect::json_extract_text.

fn case_insensitive_like(&self, col: &str, placeholder: &str) -> String

Case-insensitive LIKE comparison fragment: col <ci-like> placeholder. Postgres: col ILIKE ph. MySQL: LOWER(col) LIKE LOWER(ph). SQLite: col LIKE ph (SQLite LIKE is case-insensitive for ASCII by default).

fn sys_json_type(&self) -> &'static str

DDL fragment for a JSON/JSONB payload column (e.g. “JSONB”, “JSON”, “TEXT”).

fn sys_timestamp_type(&self) -> &'static str

Timestamp type name (without NOT NULL / DEFAULT).

fn sys_bigserial_type(&self) -> &'static str

Auto-incrementing large integer for surrogate PKs. e.g. “BIGSERIAL”, “BIGINT AUTO_INCREMENT”, “INTEGER”.

fn sys_bytes_type(&self) -> &'static str

DDL type for a raw binary payload column (e.g. “BYTEA”, “BLOB”).

fn audit_timestamp_type(&self) -> &'static str

Timestamp type used in audit table columns (no DEFAULT — values supplied explicitly).

fn supports_schemas(&self) -> bool

Whether this dialect supports CREATE SCHEMA DDL. Postgres: true. MySQL: false (uses databases). SQLite: false (no user-defined schemas).

fn default_now_plus_hours(&self, _hours: u32) -> Option<String>

DDL fragment for a column that holds a timestamp defaulting to N hours from now. Returns None when the dialect has no constant-expression equivalent (SQLite). Callers should make the column nullable and omit the DEFAULT when None is returned.

fn supports_rls(&self) -> bool

Whether this dialect natively supports row-level security (CREATE POLICY etc.).

fn supports_named_enum_types(&self) -> bool

Whether this dialect supports named enum types (CREATE TYPE … AS ENUM).

fn supports_index_include(&self) -> bool

Whether this dialect supports INCLUDE columns on indexes (Postgres 11+).

fn set_tenant_session_sql(&self, _tenant_id: &str) -> Option<String>

SQL statement that sets a session-local tenant identifier before a query. Returns None when the dialect has no such mechanism.

fn sys_timestamp_default(&self) -> String

NOT NULL timestamp column with a now() default — convenience built from above.