I am trying to get a simple elseif statement into IDL and am having a heck of a time with it. The matlab code looks something like this.
a = 1
b = 0.5
diff = a-b
thres1 = 1
thres2 = -1
if diff < thres1 & diff > thres2
'case 1'
elseif diff > thres1
'case 2'
elseif diff < thres2
'case 3'
end
But the IDL code is not so simple and I am having troubles getting the syntax right. the help states:
Syntax
IF expression THEN statement [ ELSE statement ]
or
IF expression THEN BEGIN
statements
ENDIF [ ELSE BEGIN
statements
ENDELSE ]
But doesnt give an example on how to use multiple expressions and elseif. I have tried many variations and cant seem to get it right.
Anyone have suggestions? Here are some things I've tried:
if (diff lt thres1) and (diff gt thres2) then begin
print, 'case 1'
endif else begin
if (diff gt thres1) then
print, 'case 2'
endif else begin
if (diff lt thres2) then
print, 'case 3'
endif
if (diff lt thres1) and (diff gt thres2) then begin
print, 'case 1'
else (diff gt thres1) then
print, 'case 2'
else (diff lt thres2) then
print, 'case 3'
endif
There is no elseif statement in IDL. Try:
a = 1
b = 0.5
diff = a - b
thres1 = 1
thres2 = -1
if (diff lt thres1 && diff gt thres2) then begin
print, 'case 1'
endif else if (diff gt thres1) then begin
print, 'case 2'
endif else if (diff lt thres2) then begin
print, 'case 3'
endif
So I figured it out. For those of us that are new to the IDL language.
It appears to me that IDL can only handle 2 cases for each if statement, so I had to write in another 'if' block.
hope this helps someone out there.
a = 1;
b = 2.5;
diff = a-b;
thres1 = 1;
thres2 = -1;
if diff gt thres1 then begin
print,'case 1'
endif
if (diff lt thres2) then begin
print,'case 2'
endif else begin
print,'case 3'
endelse
mgalloy's answer is correct, but you might also see people (like me), who don't use begin/endif when there's only a single line. (of course, this leads to problems if someone goes back and tries to insert a line, not realizing what you did, so Michael's approach is probably better ... this is just so that when you see this formatting, you realize it's doing the same thing:
if (diff lt thres1 && diff gt thres2) then $
print, 'case 1' $
else if (diff gt thres1) then $
print, 'case 2' $
else if (diff lt thres2) then $
print, 'case 3'
or a format that might make someone less prone to insertion:
if (diff lt thres1 && diff gt thres2) then print, 'case 1' $
else if (diff gt thres1) then print, 'case 2' $
else if (diff lt thres2) then print, 'case 3'
Related
I have a sql case statement that is working beautifully. I am trying to get the same thing to work in a Crystal report. The biggest issue is I need the report to add only the difference of two conditions, which I can't seem to make work in Crystal.
SQL:
select sum(case when type='PA' and tax_year=2022 then amt else 0 end) +
sum((case when type='UN' and tax_year in (0, 2022) then amt else 0 end) - (case when type='RA' then amt else 0 end)) -
sum(case when type='DR' and tax_year in (0,2022) then amt else 0 end) +
sum(case when type='CR' and tax_year=2022 then amt else 0 end) as total
From ar
where date_ between '7/1/2021' and '6/30/2022' and source='PT'
I have a year formula already in Crystal to total them in the appropriate year "buckets", so I am trying to get the types to work.
CRYSTAL:
If {AR.TYPE}='PA' then {AR.AMT}
else if {AR.TYPE}='DR' then -{AR.AMT}
else if {AR.TYPE}='CR' then {AR.AMT}
else if {AR.TYPE}='OP' then -{AR.AMT}
**else if {AR.TYPE}='UN' then ({AR.AMT} - if {AR.TYPE}='RA' then {AR.AMT})**
else {AR.AMT}
The bold piece above is giving me a difficult time.
I want the formula to add only the difference between all AR.TYPE='UN' and AR.TYPE='RA' but the Crystal formula I have above is just adding all UNs and ignoring the RAs
Use this logic:
If {AR.TYPE}='PA' then {AR.AMT}
else if {AR.TYPE}='DR' then -{AR.AMT}
else if {AR.TYPE}='CR' then {AR.AMT}
else if {AR.TYPE}='OP' then -{AR.AMT}
else if {AR.TYPE}='UN' then {AR.AMT}
else if {AR.TYPE}='RA' then -{AR.AMT}
else {AR.AMT}
My Table_1 looks like this
Parent_Id Child_Id Product Prod_count
1000 1 A 1
1000 2 A+B 1
1000 3 A 1
1000 4 B+C 1
2000 1 A 1
2000 2 B+C 1
2000 3 C 1
2000 4 D 1
I am trying to do Nested Loop here in this procedure,Loop based on Parent_Id, Each parent has different child so that also has to be read each row.
I have tried this
create or replace procedure sp_dummy(IN var1 int, IN var2 int, IN var3 int) as $$
Begin
create temp table find_id as(
select distinct parent_id,row_number() over(order by 1) as rw_num
from table_1
);
declare
tot_cnt int := (select count(distinct parent_id) from find_id );
init_loop int := 1;
in_init_loop int := 1;
in_tot_init_loop int;
v_parent_id int;
Begin
While init_loop <= tot_cnt
Loop
Raise info 'init_loop = %', Init_loop;
Execute 'Select parent_id into ' || v_parent_id || ' from find_id where rw_num = ' || Init_loop;
Raise info 'v_patient_id = %', v_patient_id;
Execute 'Select Count(*) into ' || in_tot_init_loop || ' from Table_1 where Parent_Id = ' || v_parent_id;
While in_init_loop <= in_tot_init_loop
Loop
Raise info 'in_init_loop = %', in_init_loop;
in_init_loop = in_init_loop + 1
End loop;
init_loop = init_loop + 1;
end loop;
End;
End;
$$ language plpgsql;
On trying this I am getting error Cannot Execute a Null Query string
I gave up on trying understanding this error!! :(
This line seems problematic:
Execute 'Select parent_id into ' || v_parent_id || ' from find_id where rw_num = ' || Init_loop;
The v_parent_id is empty, so it would translate into:
Select parent_id into NULL from find_id where rw_num = 1;
I think you actually wanted to write:
SELECT INTO v_parent_id
parent_id
FROM find_id
WHERE rw_num = Init_loop;
Yes, you can actually put the SQL in-line, rather than having to pass it as a string to EXECUTE. Take a look at the examples in Structure of PL/pgSQL - Amazon Redshift.
We get data from another company in the following formats
374-KH-ON-PEAK|807-KH-OFF-PEAK
82.5-KH-TOTAL|8-K1-CURRENT
44.5-KH-TOTAL
65-KH-ON-PEAK|2.1-K1-ON-PEAK|164-KH-OFF-PEAK|27-K1
These values go into a SQL Server table. The numbers represent electricity usages. I'm working on finding a way to extract the numbers and sum them together.
There is only one condition: the number must be followed by "-KH". If it is followed by "-K1" we don't need to do anything with it.
Upon inputting "65-KH-ON-PEAK|2.1-K1-ON-PEAK|164-KH-OFF-PEAK|27-K1", I need to output 229 which stands for 65 + 164
I'd prefer to find a solution using VBA for Access(For reasons related to the business's current software solutions), but I'm open to other solutions as well.
Using [Excel] can be done like this:
code:
Sub test()
Dim cl As Range, z!, x As Variant, x2 As Variant
For Each cl In [A1:A4]
z = 0
For Each x In Split(cl.Value2, "|")
If x Like "*-KH-*" Then
For Each x2 In Split(x, "-")
If IsNumeric(x2) Then z = z + x2
Next x2
End If
Next x
cl.Offset(, 1).Value = z
Next cl
End Sub
another variant, without second loop (using #shawnt00 comment below OP)
Sub test()
Dim cl As Range, z!, x As Variant
For Each cl In [A1:A4]
z = 0
For Each x In Split(cl.Value2, "|")
If x Like "*-KH-*" Then z = z + Left(x, InStr(1, x, "-") - 1)
Next x
cl.Offset(, 1).Value = z
Next cl
End Sub
output:
Using [Access] can be something like this:
Sub test2()
Dim z!, x As Variant
Dim rs As DAO.Recordset
Set rs = CurrentDb.OpenRecordset("SELECT * FROM Table1")
Do Until rs.EOF = True
z = 0
For Each x In Split(rs!Field1, "|")
If x Like "*-KH-*" Then z = z + Left(x, InStr(1, x, "-") - 1)
Next x
Debug.Print rs!Field1, z
rs.MoveNext
Loop
End Sub
test:
You would do a single bulk insert into an SQL Server table using | as the field terminator, so you would have fields like f1,f2,f3,f4. Then you can use an expression like:
WITH numerics
AS ( SELECT CASE
WHEN PATINDEX('%-KH-%', f1) > 0
THEN CAST(SUBSTRING(f1, 1, PATINDEX('%-KH-%', f1) - 1) AS INT)
ELSE 0
END AS f1,
CASE
WHEN PATINDEX('%-KH-%', f2) > 0
THEN CAST(SUBSTRING(f2, 1, PATINDEX('%-KH-%', f2) - 1) AS INT)
ELSE 0
END AS f2,
CASE
WHEN PATINDEX('%-KH-%', f3) > 0
THEN CAST(SUBSTRING(f3, 1, PATINDEX('%-KH-%', f3) - 1) AS INT)
ELSE 0
END AS f3,
CASE
WHEN PATINDEX('%-KH-%', f4) > 0
THEN CAST(SUBSTRING(f4, 1, PATINDEX('%-KH-%', f4) - 1) AS INT)
ELSE 0
END AS f4
FROM myTable )
SELECT f1 + f2 + f3 + f4 AS rowTotal;
You could do it with a Powershell script, that would give the power of regex to extract and sum the numbers. Something like the example below (I have tested the extracting from the file part but not the Access parts so they may need some tweaking):
$conn = New-Object -ComObject ADODB.Connection
$recordset = New-Object -ComObject ADODB.Recordset
$conn.Open()
$cmd = $conn.CreateCommand()
$ado.open("Provider = Microsoft.ACE.OLEDB.12.0;Data Source=\\path_to\database.accdb")
# Microsoft.Jet.OLEDB.4.0 for older versions of Access
(Select-String file.txt -Pattern '[\d.]+(?=-KH)' -AllMatches) | % {
($_.Matches | % {
[double]$_.Value
} | Measure-Object -Sum).Sum
} | % {
$cmd.CommandText = "INSERT INTO TABLE VALUES($($_))"
Write-Output $cmd.ExecuteNonQuery()
}
$conn.Close()
I'm making a stored procedure that loops over a table (with many thousand rows), and for each row there is a CLOB column from which I want to fetch all matches on a regular expression (sa "FNR"). Thereafter, I want to insert each distinct match in a new table. A single CLOB column may contain thousands of matches, but oftentimes it is the same "FNR" that repeats itself in the CLOB - i.e. there are much fewer distinct regex matches, and those are the only ones I'm interested in.
However, the procedure I've made takes ridiculously long time, and I suspect looping over all the matches is the most time consuming part.
My procedure looks like this:
CREATE OR REPLACE PROCEDURE SP_MTV_FINN_FNR AS
BEGIN
DECLARE
v_n NUMBER;
v_cnt NUMBER;
v_mtrid NUMBER;
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_doc CLOB;
v_fnr VARCHAR2(11);
BEGIN
-- Get all rows from table --
SELECT COUNT(*) INTO v_n FROM TABLE;
IF v_n > 0 THEN
-- Loop over all rows --
FOR i IN 1..v_n LOOP
SELECT doc, mtrid
INTO v_doc, v_mtrid
FROM (SELECT DOC doc, ID mtrid, ROWNUM rnum
FROM TABLE
WHERE ROWNUM <=i)
WHERE rnum >= i;
IF v_doc IS NOT NULL THEN
SELECT REGEXP_COUNT(v_doc, v_regex_fnr) INTO v_cnt FROM DUAL;
IF v_cnt >= 1 THEN
-- For each regex match - time consuming, right? --
FOR j IN 1..v_cnt LOOP
SELECT REGEXP_SUBSTR(v_doc, v_regex_fnr, 1, j, 'm') INTO v_fnr FROM DUAL;
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR)
SELECT v_mtrid, v_fnr FROM DUAL;
END IF;
END LOOP;
END IF;
END IF;
COMMIT;
END LOOP;
END IF;
END LOOP;
EXCEPTION WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
/
Do anybody have an idea of how to optimize this procedure?
I'm using Oracle 11.2.0.3.0.
(Btw, I know of the ctx_entity-package, but it is disabled on this version. Still, I'm thinking of enabling it.)
UPDATE
After applying the very helpful performance optimizing techniques given by nop77svk, I can with certainty say that the regexp_substr() over a CLOB is the bottleneck, as there unfortunately were no performance improvement.
However, I came up with a "hack/workaround" where I minimize the amount of regexp_substr() calls, with a tremendous performance improvement. First I thought of making an incrementally "trained" regex, excluding previous matches, but as Oracle doesn't support negative lookahead, this didn't work. I ended up with saving the CLOB, and using regexp_replace() to remove all occurences of the match. As there were a lot of the same occurences in the CLOB, this saved the procedure from a lot of regexp_substr() calls, and it simultaneously dealt with the distinct requirement.
Under follows my result, based on nop77svk's contribution. And yes, I'm back to using DUAL in the MERGE-statement, but is there any way around it here?
CREATE OR REPLACE PROCEDURE SP_MTV_FINN_FNR2 AS
BEGIN
DECLARE
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_fnr VARCHAR2(11);
v_doc CLOB;
type rec_table2 is record (
mtr_id table2.mtr_id%type,
fnr table2.fnr%type
);
type arr_table2 is table of rec_table2 index by simple_integer;
table2_bulk arr_table2;
table2_row rec_table2;
BEGIN
FOR rec IN (
select doc, MTR_ID as mtrid
from TABLE
where DOC is not null
) LOOP
v_doc := rec.doc;
loop
v_fnr := REGEXP_SUBSTR(v_doc, v_regex_fnr, 1, 1, 'm');
exit when v_fnr is null;
v_vedlegg := REGEXP_REPLACE(v_doc, v_fnr , '' , 1 , 0); -- Incrementally remove all occurences of match from doc --
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
table2_row.mtr_id := rec.mtrid;
table2_row.fnr := v_fnr;
table2_bulk(table2_bulk.count+1) := table2_row;
END IF;
END LOOP;
END LOOP;
forall i in indices of table2_bulk
MERGE INTO TABLE2 T
USING (SELECT table2_bulk(i).mtr_id mtrid, table2_bulk(i).fnr fnr FROM DUAL) B
ON (T.MTR_ID = B.mtrid AND T.FNR = B.fnr)
WHEN NOT MATCHED THEN INSERT (T.MTR_ID, T.FNR)
VALUES (B.mtrid, B.fnr);
COMMIT;
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
END;
/
Iteratively tuning your PL/SQL block ...
Iteration 0: Fixing syntax errors ...
CREATE OR REPLACE PROCEDURE SP_MTV_FINN_FNR AS
v_n NUMBER;
v_cnt NUMBER;
v_mtrid NUMBER;
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_doc CLOB;
v_fnr VARCHAR2(11);
BEGIN
-- Get all rows from table --
SELECT COUNT(*) INTO v_n FROM TABLE;
IF v_n > 0 THEN
-- Loop over all rows --
FOR i IN 1..v_n LOOP
SELECT doc, mtrid
INTO v_doc, v_mtrid
FROM (SELECT DOC doc, ID mtrid, ROWNUM rnum
FROM TABLE
WHERE ROWNUM <=i)
WHERE rnum >= i;
IF v_doc IS NOT NULL THEN
SELECT REGEXP_COUNT(v_doc, v_regex_fnr) INTO v_cnt FROM DUAL;
IF v_cnt >= 1 THEN
-- For each regex match - time consuming, right? --
FOR j IN 1..v_cnt LOOP
SELECT REGEXP_SUBSTR(v_doc, v_regex_fnr, 1, j, 'm') INTO v_fnr FROM DUAL;
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR)
SELECT v_mtrid, v_fnr FROM DUAL;
END IF;
END LOOP;
END IF;
END IF;
COMMIT;
END LOOP;
END IF;
EXCEPTION WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
Iteration 1: Removing unnecessary context switches and useless row counting ...
DECLARE
v_cnt NUMBER;
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_fnr VARCHAR2(11);
BEGIN
FOR rec IN (
select doc, id as mtrid
from table
) LOOP
IF rec.doc IS NOT NULL THEN
v_cnt := REGEXP_COUNT(rec.doc, v_regex_fnr);
IF v_cnt >= 1 THEN
-- For each regex match - time consuming, right? --
FOR j IN 1..v_cnt LOOP
v_fnr := REGEXP_SUBSTR(rec.doc, v_regex_fnr, 1, j, 'm');
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR) values (rec.mtrid, v_fnr);
END IF;
END LOOP;
END IF;
END IF;
END LOOP;
COMMIT;
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
/
Iteration 2: Decreasing number of outer loops ...
DECLARE
v_cnt NUMBER;
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_fnr VARCHAR2(11);
BEGIN
FOR rec IN (
select doc, id as mtrid
from table
where doc is not null
) LOOP
v_cnt := REGEXP_COUNT(rec.doc, v_regex_fnr);
IF v_cnt >= 1 THEN
-- For each regex match - time consuming, right? --
FOR j IN 1..v_cnt LOOP
v_fnr := REGEXP_SUBSTR(rec.doc, v_regex_fnr, 1, j, 'm');
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR) values (rec.mtrid, v_fnr);
END IF;
END LOOP;
END IF;
END LOOP;
COMMIT;
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
/
Iteration 3: Shortening the code of iteration 2 ...
DECLARE
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_fnr VARCHAR2(11);
BEGIN
FOR rec IN (
select doc, id as mtrid, REGEXP_COUNT(rec.doc, v_regex_fnr) as regexp_cnt
from table
where doc is not null
and regexp_like(doc, v_regex_fnt)
) LOOP
FOR j IN 1..rec.regexp_cnt LOOP
v_fnr := REGEXP_SUBSTR(rec.doc, v_regex_fnr, 1, j, 'm');
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR) values (rec.mtrid, v_fnr);
END IF;
END LOOP;
END LOOP;
COMMIT;
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
/
Iteration 4: Removing unnecessary regexp_count() counting ...
DECLARE
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_fnr VARCHAR2(11);
j integer;
BEGIN
FOR rec IN (
select doc, id as mtrid
from table
where doc is not null
) LOOP
j := 1;
loop
v_fnr := REGEXP_SUBSTR(rec.doc, v_regex_fnr, 1, j, 'm');
exit when v_fnt is null;
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR) values (rec.mtrid, v_fnr);
END IF;
j := j + 1;
END LOOP;
END LOOP;
COMMIT;
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
/
Iteration 5: Saving the results to memory and flushing it to DB at once (using the collection binding), plus dealing with the distinct requirement ...
create or replace type obj_table2
as
object (
mtr_id integer,
fnr varchar2(4000)
);
/
create or replace type arr_table2
as
table of obj_table2;
/
DECLARE
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_fnr VARCHAR2(11);
j integer;
table2_bulk arr_table2 := arr_table2();
BEGIN
FOR rec IN (
select doc, id as mtrid
from table
where doc is not null
) LOOP
j := 1;
loop
v_fnr := REGEXP_SUBSTR(rec.doc, v_regex_fnr, 1, j, 'm');
exit when v_fnt is null;
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
table2_bulk.extend();
table2_bulk(table2_bulk.last) := new obj_table2(
mtr_id => rec.mtrid,
fnr => v_fnr
);
END IF;
j := j + 1;
END LOOP;
END LOOP;
insert into table2(mtr_id, fnr)
select mtr_id, fnr
from table(table2_bulk) X
minus
select mtr_id, fnr
from table2;
COMMIT;
EXCEPTION
WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
/
Iteration 6: Throwing it all away whilst having decided to show off detestably ...
insert into table2 (mtr_id, fnr)
with xyz (doc, mtrid, fnr, j) as (
select doc, id as mtrid, cast(null as varchar2(4000)) as fnr, 0 as j
from table A
where doc is not null
--
union all
--
select doc, mtrid,
regexp_substr(doc, '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))', 1, j+1, 'm') as fnr,
j+1
from xyz X
where j = 0
or j > 0 and X.fnr is not null
)
select distinct mtrid, fnr
from xyz
where j > 0
and fnr is not null
and CHECK_FNR(fnr) = 'TRUE'
;
commit;
Please note that these code snippets may even not work. Since you did not provide us with any test data setup, we can tune your code only in a hypothetical way.
Please note that the slowest part of this is still the regexp_substr() over a CLOB value. You might want to think about using the position parameter of regexp_substr() instead of the occurence parameter to get the subsequent regexp matches.
Enjoy.
Removing the DUAL references with static calls. Thus saving the unwanted context switching between PL/SQL and SQL Engine.
I also added implicit cursor to process record by record.
Next level of improvement could be bulk insert into the other table. Though I didn't do it here.
CREATE OR REPLACE PROCEDURE SP_MTV_FINN_FNR AS
BEGIN
DECLARE
v_n NUMBER;
v_cnt NUMBER;
v_mtrid NUMBER;
v_regex_fnr VARCHAR2(54) := '(((0[1-9]|[12]\d|3[01])(0[1-9]|1[012])(\d{2}))(\d{5}))';
v_doc CLOB;
v_fnr VARCHAR2(11);
BEGIN
-- Get all rows from table --
/* Lets go with a Implicit cursor */
FOR MYREC IN (SELECT DOC doc, ID mtrid
FROM TABLE)
LOOP
IF MYREC.DOC IS NOT NULL THEN
v_cnt := REGEXP_COUNT(MYREC.DOC, v_regex_fnr);
IF v_cnt >= 1 THEN
-- For each regex match - time consuming, right? --
FOR j IN 1..v_cnt LOOP
v_fnr := REGEXP_SUBSTR(MYREC.DOC, v_regex_fnr, 1, j, 'm');
IF CHECK_FNR(v_fnr) = 'TRUE' THEN
INSERT INTO TABLE2(MTR_ID, FNR)
VALUES (MYREC.MTRID,v_fnr);
END IF;
END LOOP;
END IF;
COMMIT;
END IF;
END LOOP;
EXCEPTION WHEN OTHERS THEN
DBMS_OUTPUT.PUT_LINE('Error - rollback');
DBMS_OUTPUT.PUT_LINE('The error code is ' || SQLCODE || '- ' || SQLERRM);
ROLLBACK;
END;
In PL/SQL, is there any way to calculate next serial number from another one like 'A1B0010C'. Next serial no will be A1B0011C (+1). My idea is retrieve number part, increase it get return string back.
I can do this in java, but passing more than 1000 elements to Oracle will cause problems in IN clause.
So please help, any suggestion is appreciated.
Try to write some recursive function like this
This Function returns next character. ex: D after C. 0 after Z.
create or replace function get_next_character(ch char)
return char
is
result_ch char(1) := ch;
begin
IF UPPER(ch) = 'Z' or ch = '9' THEN
result_ch := 'A';
ELSE
result_ch := chr(ascii(ch) + 1);
END IF;
return upper(result_ch);
end get_next_character;
and this is your actual function which returns the next serial number
So, it generates 'A1B0010D' if your input is 'A1B0010C'
create or replace function get_next_serial(p_serial IN varchar2) -- PASS THE REQUIRED ARGUMENTS
return varchar2
IS
v_ret_serial varchar2(100);
v_in_serial varchar2(100) := p_serial;
tmp varchar2(100);
last_char char(1);
begin
tmp := v_in_serial;
for i in reverse 1..length(v_in_serial) loop
last_char := substr(tmp, length(tmp));
last_char := get_next_character(last_char);
tmp := substr(v_in_serial, 1, length(tmp)-1);
v_in_serial := substr(v_in_serial, 1, i-1) || last_char || substr(v_in_serial, i+1);
IF last_char <> 'A' then
exit;
END IF;
end loop;
IF last_char = 'A' THEN
v_in_serial:= 'A'||v_in_serial;
END IF;
return UPPER(v_in_serial);
exception
when NO_DATA_FOUND then
return 'AA';
when others then
return null;
end get_next_serial;
you can call this function (get_next_serial('abc')) where ever you want;
select get_next_serial('ALK0989KJ') from dual
You can place these two functions in a package and use at your convenience.
You can achieve it by using the following mix of Regexp_* functions:
SQL> with t1 as(
2 select 'A1B0010C' col from dual
3 )
4 select regexp_replace(col, '[[:digit:]]+'
5 , to_char(to_number(regexp_substr(col, '[[:digit:]]+',1,2) + 1), 'fm0000')
6 , 1
7 , 2
8 ) as Res
9 from t1
10 ;
RES
------------
A1B0011C
UPDATE In response to the comment.
SQL> with t1 as(
2 select 'A1B0010C' col from dual union all
3 select 'A1B0010C2' col from dual union all
4 select 'A1B0012C001' col from dual union all
5 select 'S000001' col from dual
6 )
7 select col
8 , regexp_replace(col
9 , '([[:digit:]]+)([[:alpha:]]+$|$)'
10 , LPad(to_char(to_number(num) + 1), length(num), '0') || '\2'
11 ) as Res
12 from (select col
13 , regexp_substr(col, '([[:digit:]]+)([[:alpha:]]+$|$)', 1, 1, 'i', 1) as num
14 from t1
15 )
16 ;
COL RES
----------- -----------------
A1B0010C A1B0011C
A1B0010C2 A1B0010C3
A1B0012C001 A1B0012C002
S000001 S000002
create or replace function nextSerial(s in varchar2) return varchar2 as
i integer;
done boolean := false;
c char;
newserial varchar2(4000);
begin
newserial := s;
i := length(newserial);
while i>0 and not done loop
c := substr(newserial, i, 1);
if c='9' then
newserial := substr(newserial, 1, i-1) || '0' || substr(newserial, i+1);
elsif c>='0' and c<='8' then
c := chr(ascii(c)+1);
newserial := substr(newserial, 1, i-1) || c || substr(newserial, i+1);
done := true;
end if;
i := i-1;
end loop;
if not done then
newserial := '1' || newserial;
end if;
return newserial;
end;