SQL & PL/SQL :: Hierarchical Retrieving With Cycles
Sep 29, 2011
first of all sample data;
create table test_circular_data(c1 varchar2(10),c2 varchar2(10));
insert into test_circular_data values ('c1','l2');
insert into test_circular_data values ('c1','l3');
insert into test_circular_data values ('c3','l3');
insert into test_circular_data values ('c4','l3');
commit;
There is a circular relation between columns c1 and c2, so what I'm trying to retrieve is something like that :
c1--> l2 --> l3 --> c3 --> c4
The steps to get that result is :
1.- c1 related to l2 : c1-->l2
2.- c1 related to l3 : c1-->l2-->l3
3.- l3 in the list and related to c3 : c1-->l2-->l3-->c3
4.- l3 in the list and related to c4 : c1-->l2-->l3-->c3-->c4
>select level ,empno,ename,mgr from emp connect by prior empno=mgr start with mgr is null;
Output:- > LEVEL EMPNO ENAME MGR ----- ---------- ---------- ---------- 1 7839 KING 2 7566 JONES 7839 3 7788 SCOTT 7566 4 7876 ADAMS 7788 3 7902 FORD 7566 2 7698 BLAKE 7839 3 7499 ALLEN 7698 3 7521 WARD 7698 3 7654 MARTIN 7698 3 7844 TURNER 7698 3 7900 JAMES 7698
LEVEL EMPNO ENAME MGR ----- ---------- ---------- ---------- 2 7782 CLARK 7839 3 7934 MILLER 7782
Note:- I got only this that this query is alternative of self join and it is giving manager name along with mgr id for each employee but when it gives output i couldn't able to understand how to identify who is manager of whom. Tell and explain with clause in oracle , i only know it is alternative of inline view but i want to know how does it work and how to use 'WITH' in oracle query if possible.
I want to display result with respective levels. for example p21 ,p30 are coming under first level . p22,p25 ,P31,P32are 2nd level. p23,p24,p26,p27 are 3rd level p28,p29 are FOURTH level item_id's.
Already I 'VE tried using CONNECT BY PRIOR clause.BUT STILL I COULDN'T GET THE RESULT.
I have a hierarchical data structure where a child can have many parents, and a parent can have many childs.See the attached file hierarchy_iliustration.jpg. This example has 4 hierarchy levels, in real problem there can be unlimited number of levels.I want to write a SQL query that lists all indirectly dependent child nodes for a given parent node.
Test structure for example attached:
CREATE TABLE TEST.NODE_T ( ID NUMBER PRIMARY KEY, TEXTAS VARCHAR2(254) );
I have a table which contains the connectivity details. It is telecom data. The table contains the relation between the point features(Ex: Terminal) and linear features(Ex: Cable)
I have a table (Parent - Child).There is a requirement to maintain this table, thats the hierarchy of the oraganisation.So, every quater they will be updating the table. They will be importing the data through an excel and in that excel there are 3 action items,
For Update:What all validation can come for an updation of an hierarchical data in general.Like = how to derive the level value at database side when the id is updated to some other level.= How to maintain the relation. A -> B -> D ( A is the grand parent here).A -> Ceg: if B is updated as parent node of A, then we should throw error (cyclic data). Any more validations for hierarchical data
I have a problem with my Oracle 9i SQL Query and I'm struggling to get it done.
I have three tables namely Student, Lease and Room and want to retrieve data from these three tables.
I want the Student name, the Lease details and the Room No from these tables.
The problem with my SQL query is that I get all the information from the tables except from the Room table, where in the column it show Room_No but the values are not displayed, the query is given below.
I have this SQL that returns the correct amount of rows which should be 2:
Select Distinct A.File_Name, A.File_Desc, A.file_location, A.location_date, A.downloaded_date, A.downloaded_id, A.file_size, A.days_to_request, B.File_Name, B.Act_Date, B.date_loaded from SDT_LOG A Inner Join ACTIVITY_LOG B On A.file_name = B.file_name and A.downloaded_date = B.date_loaded
I need to add another field in the Select query which is B.Act_Code. When I do, I get 2 extra rows. I do not know how to make these rows distinct.
The A table's structure is along with sample data for 1st record:
CODE Example of Name Type 1st record. ---- ------- -------------- FILE_NAME VARCHAR2(50) STLMK.txt FILE_DESC VARCHAR2(50) NON-RESIDENT FILE_LOCATION VARCHAR2(50) L:\NonResFiles YEAR NUMBER(4) 2008 LOCATION_DATE DATE 10/10/2007 DOWNLOADED_DATE DATE 09/04/2008 9:17:00 AM DOWNLOADED_ID VARCHAR2(50) Cindy FILE_SIZE CHAR(10) 16212 DAYS_TO_REQUEST NUMBER(3) 60
The B table's structure is along with sample data for 1st record:
CODE Example of Name Type 1st record ---- ------ ----------- FILE_NAME VARCHAR2(50) STLMK.txt ACT_CODE CHAR(2) D ACT_DATE DATE 10/10/2007 ACTIVITY_ID VARCHAR2(50) downloaded on DATE_LOADED DATE 09/04/2008 9:17:00 AM
The second record of activity would all be the same except Cindy would be "Jason", act_code would be an "S", activity_id would be "sent on" and then of course the dates would be changed to whenever the new information was saved within the system.
I am getting something like this (shortened of course):
CODEFile_name Downloaded_ID Act_Code
STLMK.txt Cindy D STLMK.txt Cindy S STLMK.txt Jason D STLMK.txt Jason S
There should only be one row for Cindy with a D act_code and one row for Jason with an S act_code. For some reason, Cindy and Jason each get a row with the different act_code. I'm retrieving 4 rows instead of two when I use B.Act_Code in the SQL statement.
Cindy should have the D Act_Code because she downloaded that file name and Jason should have the S because he sent that file to someone else. Every time a file's activity changes, it is entered into the system so we can keep track of where the files are.
Cindy should have the D Act_Code because she downloaded that file name and Jason should have the S because he sent that file to someone else. Every time a file's activity changes, it is entered into the system so we can keep track of where the files are.
Also, I get the 2 extra rows when I add activity_id field to the select.
First data data block contains a list item and a text item based on the selection of the list item and the value in the text item i need to retrieve more than one row in another data block whose NUMBER OF RECORDS DISPLAYED property are set to 10.
I have a Button in the first data block. So in the WHEN BUTTON PRESSED trigger i wrote SELECT INTO clause which is raising "exact fetch returns more than requested number of rows".Then i used a cursor in the WHEN-BUTTON-PRESSED trigger in the first block to fetch row by row and assign it to the items in second block. But i am able to retrieve only one record in the second block.
ID AMOUNT ID_TYPE APPROVE_FLAG A1 2 A N B1 100 B N A2 3 A N A3 100 A Y
The Select Query should be..All the records should be considered and if the (amount is greater than 50 and approve_flag is N )then except that records all records should be considered.The output should be 3 rows; 2nd row (B1) should not come. writing a select statement for the above conditions.
I have this SQL that returns the correct amount of rows which should be 2:
Select Distinct A.File_Name, A.File_Desc, A.file_location, A.location_date, A.downloaded_date, A.downloaded_id, A.file_size, A.days_to_request, B.File_Name, B.Act_Date, B.date_loaded from SDT_LOG A Inner Join ACTIVITY_LOG B On A.file_name = B.file_name and A.downloaded_date = B.date_loaded
I need to add another field in the Select query which is B.Act_Code. When I do, I get 2 extra rows. I do not know how to make these rows distinct.
The A table's structure is along with sample data for 1st record: Example of Name Type 1st record. ---- ------- -------------- FILE_NAME VARCHAR2(50) STLMK.txt FILE_DESC VARCHAR2(50) NON-RESIDENT FILE_LOCATION VARCHAR2(50) L:NonResFiles YEAR NUMBER(4) 2008 LOCATION_DATE DATE 10/10/2007 DOWNLOADED_DATE DATE 09/04/2008 9:17:00 AM DOWNLOADED_ID VARCHAR2(50) Cindy FILE_SIZE CHAR(10) 16212 DAYS_TO_REQUEST NUMBER(3) 60
The B table's structure is along with sample data for 1st record: Example of Name Type 1st record ---- ------ ----------- FILE_NAME VARCHAR2(50) STLMK.txt ACT_CODE CHAR(2) D ACT_DATE DATE 10/10/2007 ACTIVITY_ID VARCHAR2(50) downloaded on DATE_LOADED DATE 09/04/2008 9:17:00 AM
The second record of activity would all be the same except Cindy would be "Jason", act_code would be an "S", activity_id would be "sent on" and then of course the dates would be changed to whenever the new information was saved within the system.
I am getting something like this (shortened of course):
File_name Downloaded_ID Act_Code
STLMK.txt Cindy D STLMK.txt Cindy S STLMK.txt Jason D STLMK.txt Jason S
There should only be one row for Cindy with a D act_code and one row for Jason with an S act_code. For some reason, Cindy and Jason each get a row with the different act_code. I'm retrieving 4 rows instead of two when I use B.Act_Code in the SQL statement.
Cindy should have the D Act_Code because she downloaded that file name and Jason should have the S because he sent that file to someone else. Every time a file's activity changes, it is entered into the system so we can keep track of where the files are. Cindy should have the D Act_Code because she downloaded that file name and Jason should have the S because he sent that file to someone else. Every time a file's activity changes, it is entered into the system so we can keep track of where the files are.
Also, I get the 2 extra rows when I add activity_id field to the select.I use Oracle 10.
I have two blocks on which I am having relation based on booking number and employee code. As I am doing a query, booking number is getting fetched in employee code's field and employee code is getting fetched in booking number's feild. Because of this, I am not able to update the form. It is giving me error unable to update record unique contraint voilated for primary key which is quite obivious. But why the data could get shuffled in each others field though during inserting its going properly.
I have a requirement like to retrieve past 6 months data and i have used below query.
select count(1) from event where trunc(start_datae) between trunc(sydate)-180 and trunc(sysdate);it is giving the results but am not sure whether it is giving correct data or not.If ran the above query it is taking more time to execute.
Is the above approach is correct? Is there any difference between trunc(sysdate)-180 and trunc(sysdate-180) ?
How to count siblings in a hierarchical query? I'm trying to get a listing of employees
SELECT LEVEL, last_name||', '||first_name AS Manager, count(employee_id) FROM employees START WITH manager_id IS NULL CONNECT BY PRIOR employee_id = manager_id GROUP BY level
This returns 4 levels. I'm wanting to add up the number of siblings under the level 2 instead of listing them all.
I want to insert 1st two rows into Hierarchy table, then I would like to update Supervisor_2 to Supervisor_4. Here I don't want to use 'CONNECT BY PRIOR', as it take more time to execute (there are millions of records).SQL code for same.
So each Level(L1 ..L4) has zero or many child levels which further has more levels.With out using PL/SQL how can we write a Select query to give me a distinct of all children, all the way to the lowest level (L4).Example: give me all the children where L1 = 3.Result: 31, 32, 33, 311, 322, 333, 3111, 3222, 3333Is it possible to write such a query or am I asking too much logic out of a select and should go with PL/SQL.
I have a table of N records with: Name SeqNo ID Col4 ... ColX
where Name and ID are non-unique, and SeqNo is a monotonic non-consecutive sequence 0 .. N that is unique within ID..I'd like to generate the following 'groups': For each record where SeqNo = 0, sorted by Name, create the 'group where ID is the same, ordered by SeqNo irrespective of the
values of any of the other columns. For instance, if the table contained:
I got my desired results by brute-forcing via four sub-queries:
Sub-query 1 - Generate the sorted Names with SeqNo = 0 Sub-query 2 - Expand above with the additional columns, maintaining original order Sub-query 3 - For each of the records from sub-query 2, generate the 'dependents' having the same ID and SeqNo != 0 Sub-query 4 - Expand above with the additional columns, maintaining original order of sub-query 1 Main query - Create UNION of 2 and 4, sorting by original order and SeqNo
if there were not a simpler approach - after all, this must be a fairly common issue when generating BOMs.
What i want is I want the data which account is having more than 5 subaccounts within the last one month. Also need data with other date criteria like last 100 days if more than10 sub accounts. need single query.
Having more than 8lacks in my database. i wrote the query but it is taking much time and didnt give the data even after 14 hours.
My query is : select * from table a where account in (select account from table b where b.subaccount=a.subaccount and b.date>=sysdate-35 group by b.account having count(b.subaccount)>5) union select * from table a where account in (select account from table b where b.subaccount=a.subaccount and b.date>=sysdate-100 group by b.account having count(b.subaccount)>10)
how to retrieve the Subaccounts which satisfies my requirements.
I'm writing a procedure that takes a table name as a parameter and I would like to print out the column name with the supporting row entry for each row. I know the logic I'd like to use, but how do you query the metadata to return the column names and store them.
I want if the user write for example in text box 'AM TK' the query display the resualt which has am alone tk alone, and that has both. I know that i should use the Like with % but i do not know how to write it in the set property. I have wrote
set_block_property('Employee_Other',default_where,'Upper(name) like '''||UPPER(:key_search.person_name)||'''') ;
This will bring the resualt just if the user write am tk. How i can modify it to return value as i explained above.
