Extract nth substring [duplicate]

问题

This question already has answers here:

Oracle regular expression split string from last occurence (2 answers)

Split IPv4 address into 4 numbers in Oracle sql (3 answers)

Substring between characters in Oracle 9i (1 answer)

Closed last year.

I need to extract a sub string from a string. Given below are the IDs that are available.

0234-RDRT-RS111-M-EU

0234-RDRT-RSD123-M-EU

I need to extract the highlighted ones to a column. How do I extract all the characters to the same column in Microsoft SQL Server.

回答1:

The fourth parameter of REGEX_SUBSTR is the called occurence. You just have to set the occurence you want to see for each column:

CREATE TABLE T (id varchar2(30));
INSERT INTO T VALUES ('0234-RDRT-RS111-M-EU');
INSERT INTO T VALUES ('0234-RDRT-RSD123-M-EU');

SELECT regexp_substr(id,'[^-]+',1,1) as col1,
       regexp_substr(id,'[^-]+',1,2) as col2,
       regexp_substr(id,'[^-]+',1,3) as col3,
       regexp_substr(id,'[^-]+',1,4) as col4,
       regexp_substr(id,'[^-]+',1,5) as col5
  FROM t;

COL1    COL2    COL3    COL4    COL5
0234    RDRT    RS111   M   EU
0234    RDRT    RSD123  M   EU

See REGEX_SUBSTR in Oracle's documentation for more details.

回答2:

You can extract all the values of your column id by spliting them on the dash

select regexp_substr(id,'[^-]+', 1, level) 
from (select '0234-RDRT-RSD123-M-EU' as id 
        from dual)
connect by regexp_substr(id, '[^-]+', 1, level) is not null

the result has 5 lines. You can then exploit the result as needed

回答3:

Split based on '-' , store in string array then take 3rd element .

来源：https://stackoverflow.com/questions/51045592/extract-nth-substring