问题
I have written a function in DB2 - that is calculating ASCII of records in a particular column. I want to some help as I want to check the ASCII of every single character in string return yes if the ASCII of that record is greater than 127.
BEGIN
ATOMIC DECLARE POS,
INT;
IF INSTR IS NULL THEN RETURN NULL;
END IF;
SET
(
POS,
LEN
)=(
1,
LENGTH(INSTR)
);
WHILE POS <= LEN DO IF ASCII( SUBSTR( INSTR, POS, 1 ))> 128 THEN RETURN 'Y';
END IF;
SET
POS = POS + 1;
END WHILE;
RETURN 'N';
回答1:
Why to calculate ascii of every character in that column, if the goal is just to get such rows?
SELECT STR
FROM
(
VALUES
'Hello, world'
, 'Привет, мир'
) T (STR)
WHERE xmlcast(xmlquery('fn:matches($s, "[^\x00-\x7F]")' passing t.str as "s") as int) = 1;
The fn:matches function uses regular expressions.
The [^\x00-\x7F] regular expression means "a character with hex value not in the 0x00 - 0x7F interval". If a value of passed t.str contains such a character, the function returns 1 and 0 otherwise.
回答2:
A simple way to check if a UTF-8 value in DB2 only contains "plain ASCII" is to compare it's BYTE length with it's STRINGUNITS32 length. E.g.
SELECT
S
, LENGTHB(S) AS BYTES
, LENGTH4(S) AS CHARACTERS
, LENGTHB(S) = LENGTH4(S) PLAIN_ASCII
FROM
TABLE(VALUES ('123!"$'),('¹²³€½¾')) T(S)
returns
S |BYTES|CHARACTERS|PLAIN_ASCII
------|-----|----------|-----------
123!"$| 6| 6|true
¹²³€½¾| 13| 6|false
The above assumes your database is Unicode, and the columns are VARCHAR. If they are VARGRAPHIC (i.e. UTF-16), then you would need to cast them to VARCHAR in your comparision
来源：https://stackoverflow.com/questions/59473002/how-to-find-ascii-of-every-character-in-string-using-db2-function