Our approach to a dynamic where clause in PL/SQL [closed] - sql

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
We have a lot of queries in which we need to have the where clause behave dynamically, based on any amount of given filters. A filter is derived from user input (which is out of scope for this question).
We came up with the solution shown below:
A filter is represented in code by a list (plsql table) with zero or more elements, and a count variable which either contains the length of the list, or -1 if the list should be ignored (don't filter).
Example:
CREATE OR REPLACE TYPE t_number_table AS TABLE OF NUMBER;
PROCEDURE get_filter(p_filter_type IN VARCHAR2,
p_key_list OUT t_number_table,
p_count OUT PLS_INTEGER) IS
BEGIN
--normally, this procedure would consult the user input to find out what needs
--to be filtered, but for this example just always return the same filters
CASE p_filter_type
WHEN 'ORDER' THEN
p_key_list := t_number_table(11, 12, 13, 14);
p_count := 4; --user filtered on 4 orders
WHEN 'PRODUCT' THEN
p_key_list := NULL;
p_count := -1; --user did not filter on product
WHEN 'CUSTOMER' THEN
p_key_list := t_number_table(1000);
p_count := 1; --user filtered on 1 customer
END CASE;
END;
This get_filter procedure returns the filter for when the user filters on four orders, does not filter on product and filters on exactly one customer.
If we want to obtain the set of orders that matches these filters, we use it as follows:
DECLARE
v_order_list t_number_table;
v_order_count PLS_INTEGER;
v_product_list t_number_table;
v_product_count PLS_INTEGER;
v_customer_list t_number_table;
v_customer_count PLS_INTEGER;
BEGIN
get_filter('ORDER', v_order_list, v_order_count);
get_filter('PRODUCT', v_product_list, v_product_count);
get_filter('CUSTOMER', v_customer_list, v_customer_count);
...
OPEN some_ref_cursor FOR
SELECT *
FROM order_header oh
JOIN order_line ol ON ol.orderno = oh.orderno
WHERE (v_order_count = -1 OR
oh.orderno IN (SELECT * FROM TABLE(v_order_list)))
AND (v_product_count = -1 OR
ol.productno IN (SELECT * FROM TABLE(v_product_list)))
AND (v_customer_count = -1 OR
oh.customerno IN (SELECT * FROM TABLE(v_customer_list)));
END;
What we like about our approach is that it's very flexible, and somewhat elegant. It also seems to perform ok, but I'm wondering how others approach this. Also, maybe there are some issues with our approach that we missed. What are the cons against this method?

Related

what are the easiest way to create a data validation in each records in a table? [closed]

Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 1 year ago.
Improve this question
my logic on my recent project is that when user input a information thru csv , my CTL file will retrieved the data from csv to the temporary table and I want to create a data validation script to validate each data records in temporary table before transferring it to the main table.
I would likely create a procedure which grabs chunks of the temp table, checks them and then inserts them into the base table. Something like:
DECLARE
TYPE lt_toValidate IS TABLE OF temp_table%ROWTYPE INDEX BY PLS_INTEGER;
lc_fetchLimit CONSTANT INTEGER := 5000;
l_tToVal lt_toValidate;
CURSOR c_toVal IS
SELECT *
FROM temp_table;
BEGIN
OPEN c_toVal;
LOOP
FETCH c_toVal
BULK COLLECT INTO l_tToVal
LIMIT lc_fetchLimit;
EXIT WHEN l_tToVal.COUNT = 0;
FOR i IN 1..l_tToVal.COUNT LOOP
IF NOT(is_record_valid()) THEN
l_tToVal.DELETE(i);
END IF;
END LOOP;
FORALL i IN INDICES OF l_tToVal
INSERT INTO base_table (col1, col2, col3) VALUES (l_tToVal(i).col1, l_tToVal(i).col2, l_tToVal(i).col3);
END LOOP;
END;
You just need to replace is_record_valid() with whatever you are using to determine validity and obviously replace with your table and columns names.

sql project aid required [closed]

Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 6 years ago.
Improve this question
I'm trying to generate a trigger that if a column within a table is filled in with a specific set value e.g. i have a column named 'Volunteertype' which can only be assigned either 'staff' of 'student', if student is entered the trigger must nullify all columns relating to staff details within the volunteer table such as 'area of work', 'staffmanager' etc below is code i have tried to put together utilising the resources on the website however i'm having no luck.
CREATE TRIGGER trg_voltype
ON Volunteer
AFTER INSERT
AS
IF EXISTS (SELECT NULL FROM inserted i WHERE VolunteerType = 'student')
FOR EACH ROW BEGIN;
UPDATE v
SET Area_of_work = NULL
SET StaffManagerName = NULL
SET StaffManagerEmail = NULL
SET StaffManagerPhone = NULL
FROM Volunteer v
JOIN inserted i ON v.Volunteer_id = i.Volunteer_id
WHERE v.VolunteerType = 'student';
END;
However when this is run within the Oracle environment an error is produced ORA-04071: missing BEFORE, AFTER or INSTEAD OF keyword. when i attempt to shift the 'AFTER INSERT' to before the keyword i get an 'invalid trigger' error
Is anyone able to assist and inform me if the code itself is correct/ incorrect and how i should go about amending the code, thanks in advance and have a wonderful end to the year thanks!
Your trigger should be like :
CREATE OR REPLACE TRIGGER trg_voltype
BEFORE INSERT ON volunteer
FOR EACH ROW
BEGIN
if :new.Volunteertype = 'STUDENT'
then
:new.Area_of_work := NULL;
:new.StaffManagerName := NULL;
:new.StaffManagerEmail := NULL;
:new.StaffManagerPhone := NULL ;
end if;
END;

Trigger set insert row values to null [closed]

Closed. This question needs debugging details. It is not currently accepting answers.
Edit the question to include desired behavior, a specific problem or error, and the shortest code necessary to reproduce the problem. This will help others answer the question.
Closed 7 years ago.
Improve this question
So I'm trying to write a trigger that when something is inserted into the database if it is over a specific time set the row values to null
CREATE OR REPLACE TRIGGER hi
AFTER INSERT OR UPDATE OF CLASS_TIME ON class
FOR EACH ROW
BEGIN
IF (:NEW.CLASS_TIME < '09:00' )
OR (:NEW.CLASS_TIME > '18:00' )
THEN
RAISE_APPLICATION_ERROR(-20000, 'Due to low attendance no class cannot be scheduled at that time');
SET NEW.STAFFNO = NULL;
SET NEW.CLASS_DAY = NULL;
SET NEW.CLASS_TYPE = NULL;
SET NEW.ROOMNUM = NULL;
END IF;
END;
All that I have found online shows that what Ive got is correct but I get the error on the STAFFNO. Thanks in advance for any input.
First of all you have to decide:
either you want the insert or update to fail when it's being done at undesired time
or you want the insert or update to be processed, but with the columns set to null then.
You cannot have both.
In case you want to set the columns to NULL, this should be a BEFORE INSERT/UPDATE trigger, so the changed columns get written to the table. (In an AFTER INSERT/UPDATE trigger setting the fields to some value would not have any effect, becase they are not written.)
Then SET NEW.STAFFNO = NULL; is no valid PL/SQL, that would have to be :NEW.STAFFNO := NULL; instead.
CREATE OR REPLACE TRIGGER hi
BEFORE INSERT OR UPDATE OF class_time ON class
FOR EACH ROW
WHEN (new.class_time NOT BETWEEN '09:00' AND '18:00')
BEGIN
:new.staffno := null;
:new.class_day := null;
:new.class_type := null;
:new.roomnum := null;
END;

how to use triggers when comparing two data from different tables? [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 8 years ago.
Improve this question
CONSTRAINT (Movie.Date – person.DateOfBirth >= Movie.MinAge)
I am trying to use triggers for this constraint, I am not sure how to use triggers when you are comparing different data from different tables.
Something like this.. I didn't test it!!
CREATE OR REPLACE TRIGGER check_movie_parent_rating
BEFORE INSERT OR UPDATE OF DateOfBirth ON Person
REFERENCING NEW AS n
FOR EACH ROW
DECLARE
v_allowed VARCHAR2(1);
BEGIN
BEGIN
SELECT CASE WHEN (Movie.Date - person.DateOfBirth >= Movie.MinAge)
THEN 'Y'
ELSE 'N'
END As Allowed
INTO v_allowed
FROM Movie , Person
WHERE Movie.id = :n.movieId
AND Person.movieId = Movie.id
AND Person.id = :n.id;
EXCEPTION
WHEN NO_DATA_FOUND THEN
v_allowed := NULL;
END;
IF v_allowed IS NULL THEN
Raise_application_error(-20202, 'Movie Detail not available in parent');
ELSIF v_allowed = 'N' THEN
Raise_application_error(-20201, 'Movie Restricted for User');
END IF;
END;
/
Here is a solution that will avoid the mutating table error that arises when you select from a table within a row level trigger defined against the same table.
You should check the constraint whenever any involved table is changed.
Ideally, there should also be serialisation locks at within the trigger to ensure transactions executing concurrently do not alter the data in such a way as to break the constraint. This can be achieved by using modules within the supplied DBMS_LOCK package.
CREATE OR REPLACE TRIGGER Movie_Check_Age
AFTER INSERT OR UPDATE OF <Join_Columns>, Date, MinAge ON Movie
FOR EACH ROW
DECLARE
CURSOR csrPersons
IS
SELECT NULL
FROM Person p
WHERE p.<Join_Columns> = :new.<Join_Columns>
AND (Months_Between(:new.Date, p.DateOfBirth) / 12) < :new.MinAge;
rPerson csrPersons%ROWTYPE;
BEGIN
OPEN csrPersons;
FETCH csrPersons INTO rPerson;
IF csrPersons%FOUND THEN
CLOSE csrPersons;
Raise_Application_Error(-20001, 'Person too young to see movie');
ELSE
CLOSE csrPersons;
END IF;
END;
/
CREATE OR REPLACE TRIGGER Person_Check_Age
AFTER INSERT OR UPDATE OF <Join_Columns>, DateOfBirth ON Person
FOR EACH ROW
DECLARE
CURSOR csrMovies
IS
SELECT NULL
FROM Movie m
WHERE m.<Join_Columns> = :new.<Join_Columns>
WHERE (Months_Between(m.Date, :new.DateOfBirth) / 12) < m.MinAge;
rMovie csrMovies%ROWTYPE;
BEGIN
OPEN csrMovies;
FETCH csrMovies INTO rMovie;
IF csrMovies%FOUND THEN
CLOSE csrMovies;
Raise_Application_Error(-20001, 'Person too young to see movie');
ELSE
CLOSE csrMovies;
END IF;
END;
/

Conducting a COUNT without using the inbuilt COUNT(*) function [closed]

Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 9 years ago.
Improve this question
How would one go about counting fields in a table based on a condition without using the inbuilt COUNT function? Is there a way to do this?
Requirements are as follows:
Create a PL/SQL function called findtotalcarmodels to return the total
number of cars belonging to a particular model. The function should have a single
IN parameter as model_name. You should then use an explicit cursor to count the
number of cars belonging to that car model and return the final count. You must NOT
use any implicit cursors, table joins, subqueries, set operators, group functions or SQL
functions (such as COUNT) to create this function.
SUM(CASE WHEN <condition on your fields> THEN 1 ELSE 0 END)
This is probably the silliest answer I am writing. Nobody in the industry works like this.
Here goes the answer:
CREATE OR REPLACE FUNCTION findtotalcarmodels
( model_name IN VARCHAR2)
IS
CURSOR mycur IS
SELECT model_name_col FROM car_table;
mycount INTEGER := 0;
current_model_name VARCHAR2(10);
BEGIN
OPEN mycur;
FETCH mycur INTO current_model_name;
WHILE mycur%FOUND LOOP
IF current_model_name = model_name THEN
mycount := mycount + 1;
END IF;
FETCH mycur INTO current_model_name;
END LOOP;
CLOSE mycur;
dbms_output.put_line('The count is ' || mycount);
END;
/