SUBSTRING_INDEX in HANA - mysql

I'm migrating the database engine an application from MySql to SAP HANA.
I found a little trouble. I have a query like this:
Select SUBSTRING_INDEX(id, "-", -2) as prod_ref From products;
I don't know how to "translate" the function substring_index, because the initial part of the id has a variable length.
Thanks.

This can be done using a regex:
select substr_regexpr( '.*-([^-]*-[^-]*)$' in 'varia-ble---part-part1-part2' group 1) from dummy;
select substr_regexpr( '.*-([^-]*-[^-]*)$' in 'variable-part-part1-part2' group 1) from dummy;
According to the HANA 2.0 SP0 doc you could use locate with a negative offset (and then using right()), but this does not work on my system ("...feature isn't supported...")
If you execute such queries on a regular basis on lots of records I would recommend extracting the part you are interested in during ETL into a separate field. Or, alternatively fill a separate field using " GENERATED ALWAYS AS...".
I have seen it more than once, that people calculate a field like this in complex SQL queries or complex CalcViews, and then wonder why performance is bad when selecting 100 million records and filtering on the calculated field etc... Performance is usually no problem when you have aggregated your intermediate result set to a reasonable size and then apply "expensive" functions.

I don't think there is any direct function like SUBSTRING_INDEX in SAP HANA. But you have a work around alternative by creating a function to pass the input string and delimiter.
And I am assuming that -2 in SUBSTRING_INDEX and providing the solution
Reverse the string and get the position of the second delimiter, '-' in your case, into "obtainedPosition"
Now subtract that "obtainedPosition" from the length of the string.
obtainedPosition = LENGTH(id) - obtainedPosition
Using that value in the inbuilt substring function you can get the required string and return it from the function.
SELECT SCHEMA.FN_SUBSTRING_INDEX(id,obtainedPosition) INTO ReturnValue FROM DUMMY;
CREATE FUNCTION FN_SUBSTRING_INDEX
(
id VARCHAR(500),
delim VARCHAR(2)
)
RETURNS SplitString VARCHAR(500)
LANGUAGE SQLSCRIPT AS
BEGIN
DECLARE reversedString VARCHAR(500);
DECLARE charString VARCHAR(2);
DECLARE i INT := LENGTH(:id);
DECLARE len INT := LENGTH(:id);
DECLARE obtainedPosition INT := 0;
DECLARE flag INT := 0;
reversedString := '';
--loop to reverse the inputstring
WHILE :i > 0
DO
reversedString = CONCAT(:reversedString, SUBSTRING(:id,:i,1));
i := :i - 1;
END WHILE;
--loop to get the second delimiter position
i := 1;
WHILE :i <= :leng
DO
charString := '';
charString := SUBSTRING(:reversedString,i,1);
IF((:charString = :delim ) AND (:flag < 2)) THEN
BEGIN
obtainedPosition := :i;
flag := :flag + 1;
END;
END IF;
i := :i + 1;
END WHILE;
--IF condition to check if at least 2 delimiters are available, else print complete string
IF(flag = 2) THEN
obtainedPosition := :len - :obtainedPosition + 2; --2 is added to avoid the character at that position and '-' from printing
ELSE
obtainedPosition := 1;
END IF;
--SplitString contains the string's splitted return value
SELECT SUBSTRING(:id,:obtainedPosition) INTO SplitString FROM DUMMY;
END;
The above function is modified from http://www.kodyaz.com/sap-abap/sqlscript-reverse-string-function-in-sap-hana.aspx
For string functions in SAP HANA refer to this: http://www.sapstudent.com/hana/sql-string-functions-in-sap-hana/3
You can use anonymous block in SAP HANA to call and check the function
DO
BEGIN
DECLARE id VARCHAR(500) := 'Test-sam-ple-func';
DECLARE delim VARCHAR(2) := '-';
SELECT SCHEMA.FN_SUBSTRING_INDEX(id,delim) AS "SplitStringIndex" FROM DUMMY;
END;
I would be glad to know reason for a downvote. :)

Related

How to make uppercase only the odd indexes of a string in MySQL?

I'm trying to make only the odd indexes of a string in uppercase (whereas the even indexes to be in lowercase) in MySQL.
For example: StackOverflow -> StAcKoVeRfLoW or hello -> HeLlO.
I found a way to this by extracting a letter at a time using the mid function, then concatenating based on which index the letter is at:
SET #x='hello';
SELECT #x as Initial,
Concat(ucase(mid(#x,1,1)),lcase(mid(#x,2,1)),ucase(mid(#x,3,1)),lcase(mid(#x,4,1)),ucase(mid(#x,5,1)))
as Final;
However I'm interested if there is a way to simplify this, since if the string would be larger then we would have some problems. So basically is there a way to modify it to something like:
Concat(ucase(mid(#x,odd index,1)),lcase(mid(#x,even index,1)))?
This is probably most simply done in your application, but can be achieved in MySQL. For MySQL 8+ you can use a recursive CTE to extract the individual letters from the string and GROUP_CONCAT to put them back together, changing the case on an alternating basis:
WITH RECURSIVE INITIAL AS (
SELECT 'StackOverflow' AS x
),
CTE AS (
SELECT 1 AS upper, SUBSTRING(x, 1, 1) AS letter, SUBSTRING(x, 2) AS remainder
FROM INITIAL
UNION ALL
SELECT 1 - upper, SUBSTRING(remainder, 1, 1), SUBSTRING(remainder, 2)
FROM CTE
WHERE LENGTH(remainder) > 0
)
SELECT GROUP_CONCAT(CASE WHEN upper THEN UPPER(letter) ELSE LOWER(letter) END SEPARATOR '') AS new
FROM CTE
Output:
StAcKoVeRfLoW
In versions lower than 8, you can use a user-defined function:
DELIMITER //
CREATE FUNCTION AlterCase(initial TEXT)
RETURNS TEXT
DETERMINISTIC
BEGIN
DECLARE i INT DEFAULT 1;
DECLARE l CHAR(1);
DECLARE new TEXT DEFAULT '';
WHILE i <= LENGTH(initial) DO
SET l = SUBSTRING(initial, i, 1);
SET new = CONCAT(new,
CASE WHEN i % 2 = 1 THEN UPPER(l) ELSE LOWER(l) END);
SET i = i + 1;
END WHILE;
RETURN new;
END //
DELIMITER ;
And call it as
SELECT AlterCase('StackOverflow')
Output:
StAcKoVeRfLoW
Note the function will work in MySQL 8+ too.
Demo on dbfiddle

phpMyAdmin - mariaDB roman numerals function

can anybody help me with my sorting function - seriously I don't know how can I make it work as supposed to. :( Database is in MariaDB in Xampp. I use phpMyAdmin to execute the query.
DELIMITER $$
DROP FUNCTION IF EXISTS convRomanNumeral$$
CREATE FUNCTION convRomanNumeral (numeral CHAR(4))
RETURNS INT
BEGIN
DECLARE intnum INT;
CASE numeral
WHEN "I" THEN intnum = 1;
WHEN "II" THEN intnum = 2;
END CASE;
RETURN intnum;
END;
$$
SET #iteration = -1;
UPDATE `st0gk_docman_documents`
SET created_on = DATE('2016-06-14') + INTERVAL(#iteration := #iteration + 1) SECOND
WHERE `docman_category_id` = 141 ORDER BY convRomanNumeral(SUBSTRING(SUBSTRING_INDEX(title,'/',1),' ',-2) ASC, SUBSTRING_INDEX(title,'/',-2)+0 ASC;
So what I want to achieve is to sort documents by title. Example titles are:
Document Nr I/36/2006
Document Nr II/36/2006
Document Nr I/32/2006
Document Nr II/19/2006
After sorting them by first Roman number and then by second Arabic number I want to update the date. Code below for updating by only second Arabic number works properly:
SET #iteration = -1;
UPDATE `st0gk_docman_documents`
SET created_on = DATE('2016-06-14') + INTERVAL(#iteration := #iteration + 1) SECOND
WHERE `docman_category_id` = 141 ORDER BY SUBSTRING_INDEX(title,'/',-2)+0 ASC;
I would like to use CASE to return proper variable for Roman values. I know it's not perfect but I can't even make the CASE and FUNCTION work. What I am doing wrong? All suggestions are welcome.
The best way to do this is to add another column that has a sortable equivalent of that string. And use non-SQL code to do the parsing and building of that column before inserting into the table.
First mistake that I was making it was trying to execute the whole query at once... After taking the first lodge out of the way the debugging seemed way simpler. :D
So I created my case function to convert Roman numerals:
DELIMITER $$
DROP FUNCTION IF EXISTS convRomanNumeralSubFunction$$
CREATE FUNCTION convRomanNumeralSubFunction (numeral CHAR(1))
RETURNS INT
BEGIN
DECLARE intnum INT;
CASE numeral
WHEN "I" THEN SELECT 1 INTO intnum;
WHEN "X" THEN SELECT 10 INTO intnum;
WHEN "C" THEN SELECT 100 INTO intnum;
WHEN "M" THEN SELECT 1000 INTO intnum;
WHEN "V" THEN SELECT 5 INTO intnum;
WHEN "L" THEN SELECT 50 INTO intnum;
WHEN "D" THEN SELECT 500 INTO intnum;
END CASE;
RETURN intnum;
END;
$$
After that I declared the second function needed for conversion. I don't know if You can declare function inside function... and I didn't want to waste more time on this. For sure You can declare Function inside Procedure. Anyhow. WARNING: This function is not proof of BAD numerals like IIX. Numerals like that or will be badly counted. Also AXI will not count.
DELIMITER $$
DROP FUNCTION IF EXISTS convRomanNumeral$$
CREATE FUNCTION convRomanNumeral (numeral CHAR(10))
RETURNS INT
BEGIN
DECLARE currentintnum, previntnum, intnum, counter, numerallength INT;
SET numerallength = LENGTH(numeral);
SET counter = numerallength;
SET intnum = 0;
SET previntnum = 0;
WHILE counter > 0 DO
SET currentintnum = CAST(convRomanNumeralSubFunction(SUBSTRING(numeral,counter, 1)) as integer);
IF currentintnum < previntnum THEN
SET intnum = intnum - currentintnum;
ELSE
SET intnum = intnum + currentintnum;
END IF;
SET previntnum = currentintnum;
SET counter = counter - 1;
END WHILE;
RETURN intnum;
END;
$$
So that's it. Now You can convert all kind of Roman numerals and sort them up.
Use this to test the conversion:
SELECT convRomanNumeral("XIX");
This is example sorting code that I in the end used:
SET #iteration = -1;
UPDATE `st0gk_docman_documents`
SET created_on = DATE('2016-06-07') + INTERVAL(#iteration := #iteration + 1) SECOND
WHERE `docman_category_id` = 67 ORDER BY convRomanNumeralBreak(SUBSTRING_INDEX(SUBSTRING_INDEX(title,'/',1),' ',-1)) ASC, SUBSTRING_INDEX(title,'/',-2)+0 ASC;
Also one more thing - if You'll try to excecute this on mySQL then You have to fix this line:
SET currentintnum = CAST(convRomanNumeralSubFunction(SUBSTRING(numeral,counter, 1)) as integer);
into this:
SET currentintnum = CAST(convRomanNumeralSubFunction(SUBSTRING(numeral,counter, 1)) as SIGNED);
This code could be improved but as the #Rick James stated this should be done differently - not in as db update but in different table structure and sorting mechanism.

How to convert TSQL query into MYSQL query?

I have developed a function for split string in tsql but mysql don't have some built in functions. I needed to function in MYSQL as i am new in mysql. Function should accept 2 parameters
1. String to be split
2. separator (',' or whatever)
Kindly reply me.
i had found solution on the internet you can into that.
DELIMITER //
DROP FUNCTION IF EXISTS `splitAndTranslate` //
CREATE FUNCTION splitAndTranslate(str TEXT, delim VARCHAR(124))
RETURNS TEXT
DETERMINISTIC
BEGIN
DECLARE i INT DEFAULT 0; -- total number of delimiters
DECLARE ctr INT DEFAULT 0; -- counter for the loop
DECLARE str_len INT; -- string length,self explanatory
DECLARE out_str text DEFAULT ''; -- return string holder
DECLARE temp_str text DEFAULT ''; -- temporary string holder
DECLARE temp_val VARCHAR(255) DEFAULT ''; -- temporary string holder for query
-- get length
SET str_len=LENGTH(str);
SET i = (LENGTH(str)-LENGTH(REPLACE(str, delim, '')))/LENGTH(delim) + 1;
-- get total number delimeters and add 1
-- add 1 since total separated values are 1 more than the number of delimiters
-- start of while loop
WHILE(ctr<i) DO
-- add 1 to the counter, which will also be used to get the value of the string
SET ctr=ctr+1;
-- get value separated by delimiter using ctr as the index
SET temp_str = REPLACE(SUBSTRING(SUBSTRING_INDEX(str, delim, ctr), LENGTH(SUBSTRING_INDEX(str, delim,ctr - 1)) + 1), delim, '');
-- query real value and insert into temporary value holder, temp_str contains the exploded ID
SELECT <real_value_column> INTO temp_val FROM <my_table> WHERE <table_id>=temp_str;
-- concat real value into output string separated by delimiter
SET out_str=CONCAT(out_str, temp_val, ',');
END WHILE;
-- end of while loop
-- trim delimiter from end of string
SET out_str=TRIM(TRAILING delim FROM out_str);
RETURN(out_str); -- return
END//
reference http://www.slickdev.com/2008/09/15/mysql-query-real-values-from-delimiter-separated-string-ids/
In mysql they they dont support some functionality like sqlserver. so spliting will be difficult in mysql
SELECT e.`studentId`, SPLIT(",", c.`courseNames`)[e.`courseId`]
FROM ..
SELECT TRIM(SUBSTRING_INDEX(yourcolumn,',',1)), TRIM(SUBSTRING_INDEX(yourcolumn,',',-1)) FROM yourtable
CREATE FUNCTION [dbo].[SplitString]
(
#RowData nvarchar(2000),
#SplitOn nvarchar(5)
)
RETURNS #RtnValue table
(
--Id int identity(1,1),
Data nvarchar(100)
)
AS
BEGIN
Declare #Cnt int
Set #Cnt = 1
While (Charindex(#SplitOn,#RowData)>0)
Begin
Insert Into #RtnValue (data)
Select
Data = ltrim(rtrim(Substring(#RowData,1,Charindex(#SplitOn,#RowData)-1)))
Set #RowData = Substring(#RowData,Charindex(#SplitOn,#RowData)+1,len(#RowData))
Set #Cnt = #Cnt + 1
End
Insert Into #RtnValue (data)
Select Data = ltrim(rtrim(#RowData))
Return
END

How to get bitset BLOBs from Mysql select ID queries?

It would be absolutely great if the following thing was possible:
Let's say I've got a 'document' mysql table, with a document 'id' and a few other columns:
CREATE TABLE document(id INT AUTO_INCREMENT NOT NULL, ....);
There can be quite a lot of document, but let's say that for now I've got only 2 millions.
I'd like to get the result of this query in my programming language space quickly:
SELECT id FROM document WHERE ... whatever ...;
The clause 'whatever' is potentially empty, so the set can contain the id's of all documents.
So my question is: is there a way to get the result of this query as bit vector BLOB of size 2 million bits (~ 250k of data) instead of potentially 2 millions of stringified numbers ( ~ 14Mo .. not great).
Extra kudos for blob compression in case of sparse sets :)
While performance will be horrendous, this stored procedure will give you the result you requested:
CREATE PROCEDURE ex12688666(whatever TEXT)
DETERMINISTIC
READS SQL DATA
SQL SECURITY INVOKER
COMMENT ''
proc: BEGIN
DECLARE not_found BOOL DEFAULT FALSE;
DECLARE max BIGINT UNSIGNED DEFAULT 0;
DECLARE len BIGINT UNSIGNED;
DECLARE i BIGINT UNSIGNED;
DECLARE pos BIGINT UNSIGNED;
DECLARE result LONGBLOB DEFAULT '';
DECLARE cur1 CURSOR FOR
SELECT id FROM ids WHERE id RLIKE whatever ORDER BY id;
DECLARE CONTINUE HANDLER FOR NOT FOUND SET not_found = TRUE;
SELECT MAX(id) INTO max FROM ids;
IF (max > 0) THEN
SET len = FLOOR((max + 7) / 8);
SET result = REPEAT("\0", len);
OPEN cur1;
loop1: LOOP
FETCH cur1 INTO i;
IF not_found THEN
LEAVE loop1;
END IF;
SET pos = FLOOR(i / 8) + 1;
SET result = CONCAT(
SUBSTRING(result, 1, pos - 1),
CHAR(ASCII(SUBSTRING(result, pos, 1)) | (1 << (i MOD 8))),
SUBSTRING(result, pos + 1)
);
END LOOP;
CLOSE cur1;
END IF;
SELECT HEX(result) AS result;
END;
We're returning HEX(result) for illustrative purposes. In practice, one could replace
SELECT HEX(result) AS result;
with simply
SELECT result;
or if you want zlib compression of the result:
SELECT COMPRESS(result) AS result;
which should get the extra kudos you mentioned.
See http://sqlfiddle.com/#!2/6f5c0/1 for an interactive demo.

How can I pass an "array" of values to my stored procedure?

I want to be able to pass an "array" of values to my stored procedure, instead of calling "Add value" procedure serially.
Can anyone suggest a way to do it? am I missing something here?
Edit: I will be using PostgreSQL / MySQL, I haven't decided yet.
As Chris pointed, in PostgreSQL it's no problem - any base type (like int, text) has it's own array subtype, and you can also create custom types including composite ones. For example:
CREATE TYPE test as (
n int4,
m int4
);
Now you can easily create array of test:
select ARRAY[
row(1,2)::test,
row(3,4)::test,
row(5,6)::test
];
You can write a function that will multiply n*m for each item in array, and return sum of products:
CREATE OR REPLACE FUNCTION test_test(IN work_array test[]) RETURNS INT4 as $$
DECLARE
i INT4;
result INT4 := 0;
BEGIN
FOR i IN SELECT generate_subscripts( work_array, 1 ) LOOP
result := result + work_array[i].n * work_array[i].m;
END LOOP;
RETURN result;
END;
$$ language plpgsql;
and run it:
# SELECT test_test(
ARRAY[
row(1, 2)::test,
row(3,4)::test,
row(5,6)::test
]
);
test_test
-----------
44
(1 row)
If you plan to use MySQL 5.1, it is not possible to pass in an array.
See the MySQL 5.1 faq
If you plan to use PostgreSQL, it is possible look here
I don't know about passing an actual array into those engines (I work with sqlserver) but here's an idea for passing a delimited string and parsing it in your sproc with this function.
CREATE FUNCTION [dbo].[Split]
(
#ItemList NVARCHAR(4000),
#delimiter CHAR(1)
)
RETURNS #IDTable TABLE (Item VARCHAR(50))
AS
BEGIN
DECLARE #tempItemList NVARCHAR(4000)
SET #tempItemList = #ItemList
DECLARE #i INT
DECLARE #Item NVARCHAR(4000)
SET #tempItemList = REPLACE (#tempItemList, ' ', '')
SET #i = CHARINDEX(#delimiter, #tempItemList)
WHILE (LEN(#tempItemList) > 0)
BEGIN
IF #i = 0
SET #Item = #tempItemList
ELSE
SET #Item = LEFT(#tempItemList, #i - 1)
INSERT INTO #IDTable(Item) VALUES(#Item)
IF #i = 0
SET #tempItemList = ''
ELSE
SET #tempItemList = RIGHT(#tempItemList, LEN(#tempItemList) - #i)
SET #i = CHARINDEX(#delimiter, #tempItemList)
END
RETURN
END
You didn't indicate, but if you are referring to SQL server, here's one way.
And the MS support ref.
For PostgreSQL, you could do something like this:
CREATE OR REPLACE FUNCTION fnExplode(in_array anyarray) RETURNS SETOF ANYELEMENT AS
$$
SELECT ($1)[s] FROM generate_series(1,array_upper($1, 1)) AS s;
$$
LANGUAGE SQL IMMUTABLE;
Then, you could pass a delimited string to your stored procedure.
Say, param1 was an input param containing '1|2|3|4|5'
The statement:
SELECT CAST(fnExplode(string_to_array(param1, '|')) AS INTEGER);
results in a result set that can be joined or inserted.
Likewise, for MySQL, you could do something like this:
DELIMITER $$
CREATE PROCEDURE `spTest_Array`
(
v_id_arr TEXT
)
BEGIN
DECLARE v_cur_position INT;
DECLARE v_remainder TEXT;
DECLARE v_cur_string VARCHAR(255);
CREATE TEMPORARY TABLE tmp_test
(
id INT
) ENGINE=MEMORY;
SET v_remainder = v_id_arr;
SET v_cur_position = 1;
WHILE CHAR_LENGTH(v_remainder) > 0 AND v_cur_position > 0 DO
SET v_cur_position = INSTR(v_remainder, '|');
IF v_cur_position = 0 THEN
SET v_cur_string = v_remainder;
ELSE
SET v_cur_string = LEFT(v_remainder, v_cur_position - 1);
END IF;
IF TRIM(v_cur_string) != '' THEN
INSERT INTO tmp_test
(id)
VALUES
(v_cur_string);
END IF;
SET v_remainder = SUBSTRING(v_remainder, v_cur_position + 1);
END WHILE;
SELECT
id
FROM
tmp_test;
DROP TEMPORARY TABLE tmp_test;
END
$$
Then simply CALL spTest_Array('1|2|3|4|5') should produce the same result set as the above PostgreSQL query.
Thanks to JSON support in MySQL you now actually have the ability to pass an array to your MySQL stored procedure. Create a JSON_ARRAY and simply pass it as a JSON argument to your stored procedure.
Then in procedure, using MySQL's WHILE loop and MySQL's JSON "pathing" , access each of the elements in the JSON_ARRAY and do as you wish.
An example here https://gist.githubusercontent.com/jonathanvx/513066eea8cb5919b648b2453db47890/raw/22f33fdf64a2f292688edbc67392ba2ccf8da47c/json.sql
Incidently, here is how you would add the array to a function (stored-proc) call:
CallableStatement proc = null;
List<Integer> faultcd_array = Arrays.asList(1003, 1234, 5678);
//conn - your connection manager
conn = DriverManager.getConnection(connection string here);
proc = conn.prepareCall("{ ? = call procedureName(?) }");
proc.registerOutParameter(1, Types.OTHER);
//This sets-up the array
Integer[] dataFaults = faultcd_array.toArray(new Integer[faultcd_array.size()]);
java.sql.Array sqlFaultsArray = conn.createArrayOf("int4", dataFaults);
proc.setArray(2, sqlFaultsArray);
//:
//add code to retrieve cursor, use the data.
//: