I'm new in PL/SQL. I have a matrix stored in the DB as a nested table. Something like,
the matrix is stored as a TABLE of objects (and objects are t1 number, t2 number, ... t100 number)
To to get the matrix it would be select x.* from test t, table(t.matrix) x where... , returning
|T1|T2|T3|...|T100|
I want to create a function that returns the sum over the row to be called using SQL only, something equivalent to
select sum(x.T1),sum(x.T2)...sum(x.T100) from test t, table(t.matrix) x where ...
Something like select bigsum(x.*) from table t, table(t.matrix)
It will be called several times, and I don't want to write the 100 columns every time.
If you want to sum the values from 100 different columns, you're going to have to explicitly list those 100 columns at some point. You can encapsulate that logic for that expression in a view or a function or a pipelined table function or some other construct so that you don't have to repeat the expression many times, you just have to reference the abstraction you've created (i.e. call the function that sums the 100 values).
Although it would likely complicate the problem rather than simplifying it, you could potentially create a solution that uses dynamic SQL to generate the 100 columns names and the expression to add them together if you really, really want to avoid writing out 100 column names. It is highly unlikely, however, that the extra complexity of resorting to dynamic SQL would be beneficial unless there are substantial requirements that you haven't mentioned here that make writing out the column names more than a bit repetitive.
" it'll be called several times, and don't want to write the 100
columns every time"
Why not create a view? Write it once, call it as many times as you like:
create or replace view bigsum
select t.whatever
, sum(x.T1) as sum_t1
, sum(x.T2) as sum_t2
...
, sum(x.T100) as sum_t100
from test t
, table(t.matrix) x
group by t.whatever
You would need to include identifying columns from TEST to allow you to join the view to other tables. This approach would give you something close to want you want:
select *
from bigsum
where whatever = 23
You can reduce the amount of typing further by processing a result set from the data dictionary view USER_TYPE_ATTRS (or a SQL*Plus description) in a decent text editor with a regex search'n'replace.
you can create a function in the below given form depending on your condition and if you require parameter then you can add them while creating function and use them in the condition required
create or replace function bigsum
return number
as
sumall number;
begin
select (sum(x.T1),sum(x.T2)...sum(x.T100)) into sumall
from test t, table(t.matrix) x where .(your condition).. ;
return sumall;
end;/
and call it in the manner
select bigsum from dual;
Related
I have two tables, one called STUDENTS and the other CLASSES. I have to select all the students that are from the same class of one student, and this student has his own number id, and through this number id that I have to select everything.
TABLE STUDENTS
nr_rgm
nm_name
nm_father
nm_mother
dt_birth
id_sex
TABLE CLASSES
cd_class
nr_schoolyear
cd_school
cd_degree
nr_series
cd_class
cd_period
So I tried something like that :
SELECT count(*) FROM students, classes WHERE id_sex = 'M' AND
cd_class = (SELECT cd_class FROM classes WHERE nr_rgm = '12150');
But then it points an error, and the error is the follow :
single-row subquery returns more than one row
So, how can I make this work ?
you should use "in" and not "=" when applying subselects.
I think what you really would want to do is to simply join the two tables together rather than issuing a sub select:
SELECT count(*)
FROM students s, classes c
WHERE s.id_sex = 'M' AND c.nr_rgm = '12150' AND s.cd_class = c.cd_class;
This way you just tell the database: Please count all the occurrences where my students.id_sex = 'M' and my classes.nr_rgm = '12150' AND all found studends.cd_class match those of my classes.cd_class.
The reason why your statement above fails is because the ordinary = operation, when not used in a join, will only expect one single value, like you do with s.id_sex='M' while your statement returns multiple values. To cope with that you have to use the IN operator which operates on lists.
However, you can and will achieve the very same with just joining the two tables together, and it will be much more efficient on bigger data sets.
One more note to the example above. If classes.nr_rgm is a field of data type NUMBER, don't use the ' around the value 12150 as it will lead to implicit type conversion. With other words, '12150' is a string and will have to be converted to NUMBER first before doing a comparison.
I would like to have your advise how to implement the plsql. Below is the situation that i want to do..
select * from table A
loop - get each records from #1 step, and execute the store procedure, processMe(a.field1,a.field2,a.field3 || "test",a.field4);
i dont have any idea how to implement something like this. Below is sample parameter for processMe
processMe(
number_name IN VARCHAR,
location IN VARCHAR,
name_test IN VARCHAR,
gender IN VARCHAR )
Begin
select objId into obj_Id from tableUser where name = number_name ;
select locId into loc_Id from tableLoc where loc = location;
insert into tableOther(obj_id,loc_id,name_test,gender)
values (obj_Id ,loc_Id, name_test, gender)
End;
FOR rec IN (SELECT *
FROM table a)
LOOP
processMe( rec.field1,
rec.field2,
rec.field3 || 'test',
rec.field4 );
END LOOP;
does what you ask. You probably want to explicitly list the columns you actually want in the SELECT list rather than doing a SELECT * (particularly if there is an index on the four columns you actually want that could be used rather than doing a table scan or if there are columns you don't need that contain a large amount of data). Depending on the data volume, it would probably be more efficient if a version of processMe were defined that could accept collections rather than processing data on a row-by-row bases as well.
i just add some process. but this is just a sample. By the way, why
you said that this is not a good idea using loop? i interested to know
Performance wise, If you can avoid looping through a result set executing some other DMLs inside a loop, do it.
There is PL/SQL engine and there is SQL engine. Every time PL/SQL engine stumbles upon a SQL statement, whether it's a select, insert, or any other DML statement, it has to send it to the SQL engine for the execution. It calls context switching. Placing DML statement inside a loop will cause the switch(for each DML statement if there are more than one of them) as many times as many times the body of a loop has to be executed. It can be a cause of a serious performance degradation. if you have to loop, say, through a collection, use foreach loop, it minimizes context switching by executing DML statements in batches.
Luckily, your code can be rewritten as a single SQL statement, avoiding for loop entirely:
insert into tableOther(obj_id,loc_id,name_test,gender)
select tu.objId
, tl.locid
, concat(a.field3, 'test')
, a.field4
from table a
join tableUser tu
on (a.field1 = tu.name)
join tableLoc tl
on (tu.field2 = tl.loc)
You can put that insert statement into a procedure, if you want. PL/SQL will have to sent this SQL statement to the SQL engine anyway, but it will only be one call.
You can use a variable declared using a cursor rowtype. Something like this:
declare
cursor my_cursor is
select * from table;
reg my_cursor%rowtype;
begin
for reg in my_cursor loop
--
processMe(reg.field1, reg.field2, reg.field3 || "test", reg.field4);
--
end loop;
end;
I have some trouble writing SQL queries. Inside a package function, I am trying to reuse the result of a query in two other queries. Here's how it goes :
My schema stores Requests. Each Request concerns multiple destinations. Also, each Request is detailed in another table (Request_Detail). In addition, Requests are identified by their Ids.
So, I am using mainly 3 tables. One for Requests, another for the destinations and the last one for the details. Each one of theses tables is indexed by the Request_Id column.
The query I want to optimize is when a user wants to find all requests, plus their destinations and commands that have been sent between two dates.
I want to query the Request_Table first in order to get all Request_ids. Then, use this Request_Ids list to query the Command table and the Destination one.
I couldn't find how to do that... I can't use ref cursors as they can't be fetched twice... I just need some array-like or column-like variable to store the Request_Ids, then use this variable twice or more...
Here's the original queries I would like to optimize :
FUNCTION EXTRACT_REQUEST_WITH_DATE (ze_from_date DATE, ze_to_date DATE, x_request_list OUT cursor_type, x_destination_list OUT cursor_type,
x_command_list OUT cursor_type) RETURN VARCHAR2 AS
my_function_id VARCHAR2(80) := PACKAGE_ID || '.EXTRACT_REQUEST_WITH_DATE';
my_return_code VARCHAR2(2);
BEGIN
OPEN x_request_list FOR
SELECT NAME,DESTINATION_TYPE,
SUCCESS_CNT, STATUS, STATUS_DESCRIPTION,
REQUEST_ID, PARENT_REQUEST_ID, DEDUPLICATION_ID, SUBMIT_DATE, LAST_UPDATE_DATE
FROM APP_DB.REQUEST_TABLE
WHERE SUBMIT_DATE >= ze_from_date
AND SUBMIT_DATE < ze_to_date
ORDER BY REQUEST_ID;
OPEN x_destination_list FOR
SELECT REQUEST_ID, DESTINATION_ID
FROM APP_DB.DESTINATION_TABLE
WHERE SUBMIT_DATE >= ze_from_date
AND SUBMIT_DATE < ze_to_date
ORDER BY REQUEST_ID;
OPEN x_command_list FOR
SELECT SEQUENCE_NUMBER, NAME, PARAMS, DESTINATION_ID
SEND_DATE, LAST_UPDATE_DATE,PROCESS_CNT, STATUS, STATUS_DESCRIPTION,
VALIDITY_PERIOD, TO_ABORT_FLAG
FROM APP_DB.REQUEST_DETAILS_TABLE
WHERE SUBMIT_DATE >= ze_from_date
AND SUBMIT_DATE < ze_to_date
ORDER BY REQUEST_ID, DESTINATION_ID, SEQUENCE_NUMBER;
return RETURN_OK;
END EXTRACT_REQUEST_WITH_DATE;
As you see, we use the same predicate (that is the SUBMIT_DATE conditions) for all 3 queries. I think there's maybe some way to optimize it by getting REQUEST_IDs then using them in the remaining queries.
Thanks for hearing me out !
Based on the queries you posted I'd just add a SUBMIT_DATE index to REQUEST_TABLE, DESTINATION_TABLE and REQUEST_DETAILS_TABLE and leave your SQL as is. All three queries will be optimized and will run just as fast as matching against a table of REQUEST_ID values.
So...
I found this method that seems to be efficient enough :
First, defining global types to use as arrays. Here's the code :
Object(Record) type :
create or replace
TYPE "GENERIC_ID" IS OBJECT(ID VARCHAR2(64));
Variable size array of GENERIC_ID
create or replace
TYPE "GENERIC_ID_ARRAY" IS TABLE OF "GENERIC_ID";
Then, populating is done via extend() in a FOR LOOP. The resulting array can be used as a table in SQL requests, using :
TABLE(CAST(my_array_of_ids AS GENERIC_ID_ARRAY)
Thanks,
I've got an association table that groups accounts together.
I'm trying to select a subset of table 'target'
p_group_id := 7;
select *
target t
where t.account_id in get_account_ids(p_group_id);
Is it possible to write a function that returns a list of account_ids (as some form of collection) that would facilitate the above code?
I've looked at pipelined functions, however I want to stay away from loops and cursors. Also, custom types / table also get into casting that I'd like to avoid.
For reference, here's some pseudocode for what the function 'get_account_ids' would do, hypothetically:
function get_account_ids(p_group_id)
insert into v_ret
select aa.account_id
from assoc_account aa
where aa.groupid = p_group_id;
return v_ret;
You simply need:
select *
from target t
where t.account_id in
( select aa.account_id
from assoc_account aa
where aa.groupid = 7;
)
The above will work, assuming that assoc_account.account_id is never NULL.
I think I could use some help here from more experienced users...
I have an integer field name in a table, let's call it SO_ID in a table SO, and to each new row I need to calculate a new SO_ID based on the following rules
1) SO_ID consists of 6 letters where first 3 are an area code, and the last three is the sequenced number within this area.
309001
309002
309003
2) so the next new row will have a SO_ID of value
309004
3) if someone deletes the row with SO_ID value = 309002, then the next new row must recycle this value, so the next new row has got to have the SO_ID of value
309002
can anyone please provide me with either a SQL function or PL/SQL (perhaps a trigger straightaway?) function that would return the next available SO_ID I need to use ?
I reckon I could get use of keyword rownum in my sql, but the follwoing just doens't work properly
select max(so_id),max(rownum) from(
select (so_id),rownum,cast(substr(cast(so_id as varchar(6)),4,3) as int) from SO
where length(so_id)=6
and substr(cast(so_id as varchar(6)),1,3)='309'
and cast(substr(cast(so_id as varchar(6)),4,3) as int)=rownum
order by so_id
);
thank you for all your help!
This kind of logic is fraught with peril. What if two sessions calculate the same "next" value, or both try to reuse the same "deleted" value? Since your column is an integer, you'd probably be better off querying "between 309001 and 309999", but that begs the question of what happens when you hit the thousandth item in area 309?
Is it possible to make SO_ID a foreign key to another table as well as a unique key? You could pre-populate the parent table with all valid IDs (or use a function to generate them as needed), and then it would be a simple matter to select the lowest one where a child record doesn't exist.
well, we came up with this... sort of works.. concurrency is 'solved' via unique constraint
select min(lastnumber)
from
(
select so_id,so_id-LAG(so_id, 1, so_id) OVER (ORDER BY so_id) AS diff,LAG(so_id, 1, so_id) OVER (ORDER BY so_id)as lastnumber
from so_miso
where substr(cast(so_id as varchar(6)),1,3)='309'
and length(so_id)=6
order by so_id
)a
where diff>1;
Do you really need to compute & store this value at the time a row is inserted? You would normally be better off storing the area code and a date in a table and computing the SO_ID in a view, i.e.
SELECT area_code ||
LPAD( DENSE_RANK() OVER( PARTITION BY area_code
ORDER BY date_column ),
3,
'0' ) AS so_id,
<<other columns>>
FROM your_table
or having a process that runs periodically (nightly, for example) to assign the SO_ID using similar logic.
If your application is not pure sql, you could do this in application code (ie: Java code). This would be more straightforward.
If you are recycling numbers when rows are deleted, your base table must be consulted when generating the next number. "Legacy" pre-relational schemes that attempt to encode information in numbers are a pain to make airtight when numbers must be recycled after deletes, as you say yours must.
If you want to avoid having to scan your table looking for gaps, an after-delete routine must write the deleted number to a separate table in a "ReuseMe" column. The insert routine does this:
begins trans
selects next-number table for update
uses a reuseme number if available else uses the next number
clears the reuseme number if applicable or increments the next-number in the next-number table
commits trans
Ignoring the issues about concurrency, the following should give a decent start.
If 'traffic' on the table is low enough, go with locking the table in exclusive mode for the duration of the transaction.
create table blah (soc_id number(6));
insert into blah select 309000 + rownum from user_tables;
delete from blah where soc_id = 309003;
commit;
create or replace function get_next (i_soc in number) return number is
v_min number := i_soc* 1000;
v_max number := v_min + 999;
begin
lock table blah in exclusive mode;
select min(rn) into v_min
from
(select rownum rn from dual connect by level <= 999
minus
select to_number(substr(soc_id,4))
from blah
where soc_id between v_min and v_max);
return v_min;
end;