WebAssembly ユーザー定義関数

<clickhouse>
    <allow_experimental_webassembly_udf>true</allow_experimental_webassembly_udf>
    <webassembly_udf_engine>wasmtime</webassembly_udf_engine>
</clickhouse>

cat << 'EOF' | wasm-tools parse | clickhouse client -q "INSERT INTO system.webassembly_modules (name, code) SELECT 'collatz', code FROM input('code String') FORMAT RawBlob"
(module
  (func $next (param $n i32) (result i32)
    local.get $n i32.const 1 i32.and
    (if (result i32)
      (then local.get $n i32.const 3 i32.mul i32.const 1 i32.add)
      (else local.get $n i32.const 2 i32.div_u)))
  (func $steps (export "steps") (param $n i32) (result i32)
    (local $count i32)
    local.get $n i32.const 1 i32.lt_u
    (if (then i32.const 0 return))
    (block $done (loop $loop
      local.get $n i32.const 1 i32.eq br_if $done
      local.get $n call $next local.set $n
      local.get $count i32.const 1 i32.add local.set $count
      br $loop))
    local.get $count)
)
EOF

CREATE FUNCTION collatz_steps LANGUAGE WASM ARGUMENTS (n UInt32) RETURNS UInt32 FROM 'collatz' :: 'steps';

SELECT groupArray(collatz_steps(number :: UInt32))
FROM numbers(1, 100)
FORMAT TSV

[0,1,7,2,5,8,16,3,19,6,14,9,9,17,17,4,12,20,20,7,7,15,15,10,23,10,111,18,18,18,106,5,26,13,13,21,21,21,34,8,109,8,29,16,16,16,104,11,24,24,24,11,11,112,112,19,32,19,32,19,19,107,107,6,27,27,27,14,14,14,102,22,115,22,14,22,22,35,35,9,22,110,110,9,9,30,30,17,30,17,92,17,17,105,105,12,118,25,25,25]

INSERT INTO system.webassembly_modules (name, code)
SELECT 'my_module', base64Decode('AGFzbQEAAAA...');

INSERT INTO system.webassembly_modules (name, code, hash)
SELECT 'my_module', base64Decode('...'), reinterpretAsUInt256(unhex('369f...c57d'));

SELECT name, lower(hex(reinterpretAsFixedString(hash))) AS sha256 FROM system.webassembly_modules

   ┌─name────┬─sha256───────────────────────────────────────────────────────────┐
1. │ collatz │ a084a10b7b5cb07db198bc93bf1f3c1f8cb8ef279df7a4f6b66b1cdd55d79c48 │
   └─────────┴──────────────────────────────────────────────────────────────────┘

DELETE FROM system.webassembly_modules WHERE name = 'collatz';

-- 名前が `tmp_` で始まるすべてのモジュールを一括削除する（リテラルのアンダースコアは `\_` としてエスケープされる）:
DELETE FROM system.webassembly_modules WHERE name LIKE 'tmp\_%';

CREATE [OR REPLACE] FUNCTION function_name
LANGUAGE WASM
FROM 'module_name' [:: 'source_function_name']
ARGUMENTS ( [name type[, ...]] | [type[, ...]] )
RETURNS return_type
[ABI ROW_DIRECT | ABI BUFFERED_V1]
[DETERMINISTIC]
[SHA256_HASH 'hex']
[SETTINGS key = value[, ...]];

(func (param i32 i64 f32) (result f64) ...)

CREATE FUNCTION my_func ARGUMENTS (Int32, UInt64, Float32) RETURNS Float64 ...

(module
  ;; 指定されたサイズの新しいバッファを割り当てる
  ;; 戻り値: Buffer構造体へのハンドル（データへの直接ポインタではない）。データへのポインタとサイズを含む
  (func (export "clickhouse_create_buffer")
    (param $size i32)    ;; 割り当てるデータのサイズ
    (result i32))        ;; 十分な領域を持つバッファハンドルを返す

  ;; ハンドルを使ってバッファを解放する
  (func (export "clickhouse_destroy_buffer")
    (param $handle i32)  ;; 解放するバッファハンドル
    (result))            ;; 戻り値なし

    ;; ユーザー定義関数
    (func (export "user_defined_function1")
      (param $input_buffer_handle i32)  ;; 入力バッファハンドル
      (param $n i32)                    ;; 入力の行数
      (result i32))                     ;; 出力バッファハンドルを返す
)

typedef struct {
    uint8_t * data;
    uint32_t size;
} ClickhouseBuffer;

ClickhouseBuffer * clickhouse_create_buffer(uint32_t size) { /* ... */ }

void clickhouse_destroy_buffer(ClickhouseBuffer * data) { /* ... */ }

/// ユーザー定義関数の例
ClickhouseBuffer * user_defined_function1(ClickhouseBuffer * span, uint32_t n) { /* ... */ }
ClickhouseBuffer * user_defined_function2(ClickhouseBuffer * span, uint32_t n) { /* ... */ }

use clickhouse_wasm_udf_bindgen::clickhouse_udf;

#[clickhouse_udf]
pub fn some_udf(data: String) -> HashMap<String, String> {
    // ここに実装を記述してください
}

SET webassembly_udf_max_fuel = 200000;
SELECT my_wasm_udf(column) FROM table;