PostgreSQL中的Base58编码器函数

Question

我发现这个MySQL函数是为了

Base58

中的编码器

Github要点

..。

DELIMITER $$

CREATE FUNCTION base58_encode (num int) RETURNS varchar(255)
  DETERMINISTIC

BEGIN
  DECLARE alphabet varchar(255);
  DECLARE base_count int DEFAULT 0;
  DECLARE encoded varchar(255);
  DECLARE divisor DECIMAL(10,4);
  DECLARE mode int DEFAULT 0;

  SET alphabet = '123456789abcdefghijkmnopqrstuvwxyzABCDEFGHJKLMNPQRSTUVWXYZ';
  SET base_count = CHAR_LENGTH(alphabet);
  SET encoded = "";

  WHILE num  >= base_count DO
    SET divisor = num / base_count;
    SET mode = (num - (base_count* TRUNCATE(divisor,0)));
    SET encoded = CONCAT(SUBSTRING(alphabet FROM mode+1 FOR 1), encoded);
    SET num = TRUNCATE(divisor,0);
  END WHILE;

  SET encoded = CONCAT(SUBSTRING(alphabet FROM num+1 FOR 1), encoded);

  RETURN (encoded);

END

我是新手

并且难以将上述函数转换为PostgreSQL函数。

怎么会有等价物呢？

以上SQL代码片段在Base58编码器中的作用？

Answer 1

我在PostgreSQL中想出的等价函数如下所示。

CREATE FUNCTION base58_encode(num INT)
  RETURNS VARCHAR(255) AS $encoded$

DECLARE
  alphabet   VARCHAR(255);
  base_count INT DEFAULT 0;
  encoded    VARCHAR(255);
  divisor    DECIMAL(10, 4);
  mod        INT DEFAULT 0;

BEGIN
  alphabet := '123456789abcdefghijkmnopqrstuvwxyzABCDEFGHJKLMNPQRSTUVWXYZ';
  base_count := char_length(alphabet);
  encoded := '';

  WHILE num >= base_count LOOP
    divisor := num / base_count;
    mod := (num - (base_count * trunc(divisor, 0)));
    encoded := concat(substring(alphabet FROM mod + 1 FOR 1), encoded);
    num := trunc(divisor, 0);
  END LOOP;

  encoded = concat(substring(alphabet FROM num + 1 FOR 1), encoded);

  RETURN (encoded);

END; $encoded$
LANGUAGE PLPGSQL;

Answer 2

为了完整起见，下面是对反面的快速粗略描述

功能：

CREATE OR REPLACE FUNCTION base58_decode(str VARCHAR(255))
  RETURNS BIGINT AS $$
DECLARE
  alphabet VARCHAR(255);
  c CHAR(1);
  p INT;
  v BIGINT;
BEGIN
  alphabet := '123456789abcdefghijkmnopqrstuvwxyzABCDEFGHJKLMNPQRSTUVWXYZ';
  v := 0;
  FOR i IN 1..char_length(str) LOOP
    c := substring(str FROM i FOR 1);
    -- This is probably wildly inefficient, but we're just using this function for diagnostics...
    p := position(c IN alphabet);
    IF p = 0 THEN
      RAISE 'Illegal base58 character ''%'' in ''%''', c, str;
    END IF;
    v := (v * 58) + (p - 1);
  END LOOP;
  RETURN v;
END;$$
LANGUAGE PLPGSQL;

Answer 3

Postgres 9.x

CREATE OR REPLACE FUNCTION base58_encode (num bigint)
RETURNS text AS
$body$
declare
  --alphabet text = '123456789abcdefghijkmnopqrstuvwxyzABCDEFGHJKLMNPQRSTUVWXYZ';
  alphabet text[] = array[
    '1','2','3','4','5','6','7','8','9', 
    'a','b','c','d','e','f','g','h','i','j','k','m','n','o','p','q','r','s','t','u','v','w','x','y','z',
    'A','B','C','D','E','F','G','H','J','K','L','M','N','P','Q','R','S','T','U','V','W','X','Y','Z'
  ];
  cnt integer = 58;
  dst text = '';
  mod integer;
begin
  while (num >= cnt) loop
    num = num / cnt;
    mod = num % cnt + 1;
    dst = alphabet[mod] || dst;
  end loop;

  return alphabet[num] || dst;
end;
$body$
LANGUAGE 'plpgsql'
IMMUTABLE
CALLED ON NULL INPUT
SECURITY INVOKER
COST 100;