Struct LoadFileParams 

pub struct LoadFileParams {
    pub table: String,
    pub path: String,
    pub mode: Option<String>,
    pub schema: Option<Value>,
    pub json_extract_path: Option<String>,
    pub merge_key: Option<MergeKey>,
    pub database: Option<String>,
    pub persist: Option<bool>,
}

Parameters for the load_file workspace tool.

Fields

table: String

Target table name.

path: String

Absolute path to a CSV, Parquet, or Arrow IPC file.

mode: Option<String>

"replace" (default — drops and recreates the table), "append" (adds rows to an existing table), or "merge" (upserts rows by merge_key; new columns in the incoming file are auto-added via ALTER TABLE ADD COLUMN).

schema: Option<Value>

Partial schema override keyed by column name: {"col": "BIGINT", ...}. Only the listed columns are overridden; the rest keep their inferred type. Call inspect_file first if you are unsure — it reports min / max / null_count per column using the exact same inference this tool uses, so the override you build from its output is guaranteed to align with the file’s actual columns.

json_extract_path: Option<String>

Optional dot-separated path to extract a nested data array from the JSON file. Numeric segments index into arrays (e.g., content.0). String values encountered during navigation are automatically parsed as JSON, handling the common pattern where MCP tool responses contain stringified JSON payloads.

merge_key: Option<MergeKey>

When mode = "merge", the column(s) used to match incoming rows to existing rows for upsert. Pass a single name ("job_id") or a list (["cell", "job_id"]). Required for merge; rejected with a clear error if set for replace or append.

database: Option<String>

Target database alias. Omit (or pass "local") to write to the ephemeral primary. Pass "persistent" to write to the durable database. Other values target a user-attached writable database.

persist: Option<bool>

Shorthand for database: "persistent". If both database and persist are set, database wins.

Trait Implementations

impl Debug for LoadFileParams

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for LoadFileParams

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for LoadFileParams

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.