Hierarchical Queries - Challenging

Hi all,
Thanks in advance,
please reply as soon as possible
Let me know some suggestions to get the non duplicate PRDCT_CUSTM_HIER_SK values using Hierarchical Querie of oracle 10g
suggest on how to edit the query without changing the CONNECT BY PRIOR condition
On only one condition the query should never return duplicate PRDCT_CUSTM_HIER_SK values.
If you view the example we have 4 duplicate values
consider the table as below
PRDCT_CUSTM_HIER_SK -- ITEM_ID -- PRNT_ID
27945 -- 74 -- 0
28977 -- 75 -- 74
28100 -- 85 -- 75
28538 -- 86 -- 85
29557 -- 115 -- 74
29477 -- 116 -- 115
28117 -- 360 -- 86
30894 -- 360 -- 85
29039 -- 390 -- 115
30876 -- 431 -- 360
30839 -- 432 -- 360
30883 -- 433 -- 360
30863 -- 434 -- 360
I HAVE EXECUTED THE GIVEN SCRIPT
SELECT
LEVEL CUSTM_HIER_LVL_ID
,ITEM_ID CUSTM_HIER_ITEM_ID
,PRNT_ID CUSTM_HIER_PRNT_ID
,lpad(PRDCT_CUSTM_HIER_SK,length(PRDCT_CUSTM_HIER_SK) + level *10 - 10,'-') PRDCT_CUSTM_HIER_SK
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),2,'/')) CUSTM_HIER_SK_LVL1
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),3,'/')) CUSTM_HIER_SK_LVL2
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),4,'/')) CUSTM_HIER_SK_LVL3
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),5,'/')) CUSTM_HIER_SK_LVL4
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),6,'/')) CUSTM_HIER_SK_LVL5
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),7,'/')) CUSTM_HIER_SK_LVL6
,TO_NUMBER(GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_SK,'/'),8,'/')) CUSTM_HIER_SK_LVL7
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),2,'/') CUSTM_HIER_NM_LVL1
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),3,'/') CUSTM_HIER_NM_LVL2
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),4,'/') CUSTM_HIER_NM_LVL3
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),5,'/') CUSTM_HIER_NM_LVL4
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),6,'/') CUSTM_HIER_NM_LVL5
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),7,'/') CUSTM_HIER_NM_LVL6
,GET_TOKEN(SYS_CONNECT_BY_PATH(PRDCT_CUSTM_HIER_LVL_1_NM,'/'),8,'/') CUSTM_HIER_NM_LVL7
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),2,'/') CUSTM_HIER_ITEM_ID_LVL1
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),3,'/') CUSTM_HIER_ITEM_ID_LVL2
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),4,'/') CUSTM_HIER_ITEM_ID_LVL3
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),5,'/') CUSTM_HIER_ITEM_ID_LVL4
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),6,'/') CUSTM_HIER_ITEM_ID_LVL5
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),7,'/') CUSTM_HIER_ITEM_ID_LVL6
,GET_TOKEN(SYS_CONNECT_BY_PATH(ITEM_ID,'/'),8,'/') CUSTM_HIER_ITEM_ID_LVL7
,CONNECT_BY_ROOT PRDCT_CUSTM_HIER_SK PRDCT_CUSTM_ROOT_SK
,PRIOR PRDCT_CUSTM_HIER_SK PRDCT_CUSTM_PRNT_SK
,PRDCT_CUSTM_HIER_LVL_1_NM PRDCT_CUSTM_HIER_NM
,PRIOR PRDCT_CUSTM_HIER_LVL_1_NM CUSTM_HIER_PRNT_NM
,CONNECT_BY_ROOT PRDCT_CUSTM_HIER_LVL_1_NM CUSTM_HIER_ROOT_NM
,SYS_CONNECT_BY_PATH (PRDCT_CUSTM_HIER_LVL_1_NM, '/') CUSTM_HIER_NM_PATH
,PRDCT_CUSTM_HIER_SS_CD PRDCT_CUSTM_HIER_SS_CD
,PRDCT_BRND_NM PRDCT_BRND_NM
,PRDCT_BRND_ID PRDCT_BRND_ID
,PRDCT_BRND_SK PRDCT_BRND_SK
FROM
BI.TMP_PRDCT_CUSTM_HIER
START WITH ITEM_ID = (SELECT MIN(ITEM_ID) FROM BI.TMP_PRDCT_CUSTM_HIER)
CONNECT BY PRIOR ITEM_ID = PRNT_ID
THE OUTPUT IS AS FOLLOWS
1 -- 74 -- 0 -- 27945
2 -- 115 -- 74 -- ----------29557
3 -- 390 -- 115 -- --------------------29039
3 -- 116 -- 115 -- --------------------29477
2 -- 75 -- 74 -- ----------28977
3 -- 85 -- 75 -- --------------------28100
4 -- 360 -- 85 -- ------------------------------30894
*5 -- 431 -- 360 -- ----------------------------------------30876*
*5 -- 432 -- 360 -- ----------------------------------------30839*
*5 -- 433 -- 360 -- ----------------------------------------30883*
*5 -- 434 -- 360 -- ----------------------------------------30863*
4 -- 86 -- 85 -- ------------------------------28538
5 -- 360 -- 86 -- ----------------------------------------28117
*6 -- 431 -- 360 -- --------------------------------------------------30876*
*6 -- 432 -- 360 -- --------------------------------------------------30839*
*6 -- 433 -- 360 -- --------------------------------------------------30883*
*6 -- 434 -- 360 -- --------------------------------------------------30863*

Hi,
Since item_id 360 has two parents (item_ids 85 and 86),, then 360 and all of its descendants will appear twice in the output.
If you want them to appear only once, you have to tell the query to only count one of the two parents of 360.
Here's one way to ignore all but one parent of any item:
WITH     got_r_num     AS
     SELECT     product_custm_hier_sk
     ,     item_id
     ,     print_id
     ,     ROW_NUMBER () OVER ( PARTITION BY item_id
                               ORDER BY         prnt_id
                       )      AS r_num
     FROM     bi.tmp_prdct_custm_hier
--     WHERE     ...     -- any filtering goes here
,     one_parent     AS
     SELECT     *
     FROM     got_r_num
     WHERE     r_num     = 1
SELECT     ...          -- Continue with your query
FROM     one_parent     -- only change to main query
START WITH     ...That is, instead of using the actual table in the FROM clause, you use the result set of a Top-N query.
If you don't always want the lowest parent to be the one used, then just change the analytic ORDER BY clause.
If you need help, post a little sample data (CREATE TABLE and INSERT statements) and the results you want from that data.
When posting output (or any formatted text) on this site, type these 6 characters:
\(small letters only, inside curly brackets) before and after each section of formatted text, to preserve the spacing.

Similar Messages

Help needed in understanding the concept of hierarchical queries

I really need help in this matter. I have a flafile containing about 4000 rows. It is from my supplier, it's structure is as follows:
create table Flatfile
(Pgroup varchar2(30),
Pclass varchar2(30),
Manufacturer varchar2(30),
Article varchar2(30),
Price Number(6,2));
Insert into Flatfile Values
('Application Software','Database Software','Oracle','Oracle 10G',115);
Insert into Flatfile Values
('Application Software','Database Software','Microsoft','MS SQL Server 2000',200);
Insert into Flatfile Values
('Application Software','Spreadsheet Software','Microsoft','Excel',100);
Insert into Flatfile Values
('Monitor','15"','Acer','Acer 15"" TFT superscreen',199);
Insert into Flatfile Values
('Monitor','15"','Sony','Sony R1500 flat',225);
Insert into Flatfile Values
('Monitor','17"','Philips','Philips Flatscreen',250);
Insert into Flatfile Values
('Monitor','19"','Viewsonic','Viewsonic PLasma Monitor',275);
Insert into Flatfile Values
('Processor','AMD','AMD','FX-55',600);
Insert into Flatfile Values
('Processor','Intel','Intel','P4 3 GHZ',399);
My goal is to make a hierarchical query with the start with and connect by clauses. From what I have read is that I need to normalize the data of the flatfile.
How do I achieve a table which I can query so that the query will represent the hierarchy that exists. Namely
Pgroup
++Pclasse
Application Software
++Database Software
++Spreadsheet Software
So a 2-level hierarchy. I'd like to understand this simple concept first. I built on the knowledge that I gain. So the questions are:
1.What do I need to do to make the table so that I can use a hierarchical query on it?
2. How should the query syntax be?
3. Is it also possible to get the data in the hierarchical query sorted asec?
I would only like to use the simple structures of the start with and connect by clauses first. I've read there are some new additions to 10G. The problem with the examples used by the tutorials is that the tables are already made so that they are suitable for hierarchical queries. I hope to understand it by this example. And take it a step further.
Sincerely,
Pete

Primarily hierarchy query serves to process tree-like structures which RDBMS simulates using through parent-child relation, often in a single table (see famoust
EMP table where employee can have the manager who is an employee at the same time).
In your case it could look like:
SQL> select pgroup, pclass from flatfile;
PGROUP                         PCLASS
Application Software           Database Software
Application Software           Database Software
Application Software           Spreadsheet Software
Monitor                        15"
Monitor                        15"
Monitor                        17"
Monitor                        19"
Processor                      AMD
Processor                      Intel
                               Application Software
                               Monitor
                               Processor
12 rows selected.
SQL> select decode(level,1,pclass,' ' || pclass), Manufacturer from flatfile
2 start with pgroup is null
3 connect by prior pclass = pgroup
4 /
DECODE(LEVEL,1,PCLASS,''||PCLASS MANUFACTURER
Application Software
Database Software              Oracle
Database Software              Microsoft
Spreadsheet Software           Microsoft
Monitor
15"                            Acer
15"                            Sony
17"                            Philips
19"                            Viewsonic
Processor
AMD                            AMD
Intel                          Intel
12 rows selected.The hierarchy syntax is described completely in the documentation including
LEVEL and PRIOR keywords.
As for the ordering question you can use siblings ordering:
SQL> select decode(level,1,pclass,' ' || pclass), Manufacturer from flatfile
2 start with pgroup is null
3 connect by prior pclass = pgroup
4 order siblings by 1 desc
5 /
DECODE(LEVEL,1,PCLASS,''||PCLASS MANUFACTURER
Processor
Intel                          Intel
AMD                            AMD
Monitor
19"                            Viewsonic
17"                            Philips
15"                            Acer
15"                            Sony
Application Software
Spreadsheet Software           Microsoft
Database Software              Oracle
Database Software              Microsoft
12 rows selected.Rgds.

Alternative for Hierarchical Queries

Hi all,
Is there any other way to implement the Hierachical Query in Oracle. Let us assume the following example of the Scott.emp Table. The output of the table must be in a Hierarchical manner as follows :
ORG_CHART EMPNO MGR JOB
KING 7839 PRESIDENT
JONES 7566 7839 MANAGER
SCOTT 7788 7566 ANALYST
ADAMS 7876 7788 CLERK
FORD 7902 7566 ANALYST
SMITH 7369 7902 CLERK
BLAKE 7698 7839 MANAGER
ALLEN 7499 7698 SALESMAN
WARD 7521 7698 SALESMAN
MARTIN 7654 7698 SALESMAN
TURNER 7844 7698 SALESMAN
JAMES 7900 7698 CLERK
CLARK 7782 7839 MANAGER
MILLER 7934 7782 CLERK
The above structure can be achieved using the following implementation by using the clauses namely CONNECT BY PRIOR, LEVEL and START WITH :
SELECT LPAD(' ',2*(LEVEL-1)) || ename org_chart,
empno, mgr, job
FROM emp
START WITH job = 'PRESIDENT'
CONNECT BY PRIOR empno = mgr;
The above query works fine without any issues.
But is there any other way to implement the above logic without using the above hierarchical query clauses.
Please help me on the above.
Thanks in advance.
Regards
Raj

Thanks.
Why I require this implementation is we have an software which runs both on Oracle and SQL Server, we accomplish the same very easily in Oracle by using the Hierarchical Queries already available in Oracle. But there is no such predefined keywords to implement in SQL server. That is the purpose of the above.
When we can achieve the same in a alternative way in Oracle, the same I feel can be implemented in SQL server.
Raj
Not without dropping into PLSQL, but this is not easy, and why bother, if your query works OK?

Hierarchical Queries

I've a table with these values.
ID     Name          Parent_ID
0     Organization1     null
1     Organization2     0
2     Organization3     0
3     Organization4     0
4     Organization5     1
5     Organization6     1
6     Organization7     2
7     Organization8     2
8     Organization9     4
9     Organization10     4
10     Organization11     6
11     Organization12     8.
Need to write a sql query in Oracle 9i to get the Hierarchy of Organization12:
The output should be:
Organization1
Organization9
Organization5
Organization2
Organization1.
The query should be dynamic. If I give Organization11, I should get the hierarchy of Organization11.
Organization1
Organization7
Organization3
Organization1.
Could anyone please help me in getting this output.

See the followings links for hierarchical queries:
http://philip.greenspun.com/sql/trees.html
http://www.psoug.org/reference/connectby.html
--Manohar

Map the hierarchical queries pseudo column level

Is it possible to map the hierarchical queries pseudo column level in the Workbench?
I have an object for which I use connect by queries. I would like to map the pseudo column level. is this possible? if yes, how
If the mapping is not possible how do I get the level using a report query?
Is there any other way to retrive level?
Thanks
Edited by: amehta5 on May 4, 2010 11:52 AM
Edited by: amehta5 on May 4, 2010 11:55 AM

Thanks James, appreciate your feedback.
I tried report.addItem("level", builder.getFunction("LEVEL"));
but the query TopLink generates has LEVEL() and it errors out.
Code -
Expression startExpr = null;
Expression connectBy = builder.get("manager");
Vector<Expression> order = new Vector<Expression>();
order.addElement(builder.get("name"));
report.setHierarchicalQueryClause(startExpr, connectBy, order);
report.addAttribute("name");
report.addItem("level", builder.getFunction("LEVEL"));
Query generated by TOpLink - SELECT NAME, ID, LEVEL() FROM EMPLOYEE WHERE (EMPLOYEE_TYPE = ?) CONNECT BY PRIOR EMPLOYEE.MANAGER_ID = EMPLOYEE.ID ORDER SIBLINGS BY NAME
bind => [M]
Edited by: amehta5 on May 6, 2010 6:11 AM

Hierarchical queries (mother, father,...)

Hi guys,
a have a problem that I think it can be solved whit hierarchical queries.
the question is like this. I have a person ...and each person has a mother and a father. (the mother/father may be or not be in my table)
so my table configuration will be like:
create table person
(id number,
name varchar2 (100),
mother_id number,
father_id number
the questions I need to answer are like:
1. return all persons that have all grandparents (or grand grand parents) in this table.(including parents). so entire genealogic tree is intact
e.g. if the grandfather is not in the table I will not display that person we cannot display that person
I have something like:
insert into person values (1,'alin', 4, 5); ----id 1 has parents in my table: 4 the mother, 5 the father
insert into person values (4,'mother_1', 8,null); ----id 4 has only her mother in table
insert into person values (5, 'father_1',9,10);--id 5(who is father of 1) with 9 as mother and 10 as father
insert into person values (8, 'grant',null ,null );
insert into person values (9, 'aaa',11,12);
insert into person values (10,'bbb',13,14);
insert into person values (11,'ccc',null ,null);
insert into person values (12, 'ddd',null ,null);
insert into person values (13,'eee',15,16);
insert into person values (14,'fff',17,18);
insert into person values (15,'ggg',null ,null);
insert into person values (16,'hhh',null ,null);
insert into person values (17,'iii',null ,null);
insert into person values (18,'jjjj',null,null);
in this configuration my select should return id 5 with his relatives and id 10 with this relatives because only them have all relatives in table:
id         |         name      |          mother    |      father      |      mather_mother (mm)   |        father_mother (fm) |      mf    |        ff
5          |         'father'    |           9           |      10      |              11            |            12              |     13     |      14
10        |          bbb        |           13         |      14      |                15            |             16              |       17   |        18
and second select ..if I input an id..lets say 1 to get the same result..like:
id        |          name         |       mother     |     father   |         mather_mother (mm) |         father_mother (fm) |       mf   |         ff
1         |         'alin'           |      4              |   5           |             8           |       null           |      9       |     10and maybe at some point I want to show level 3(with grand grand parents)
please se the attached picture:
link: [http://picasaweb.google.com/alinbor/Tree#5428860555382426514]
Thanks

Hi,
Here's a revised version of the String Aggregation approach.
For :level_cnt = 3, it produces this output:
   ROOT_ID ROOT_NAME ANCESTORS
                          M    F   MM   FM   MF   FF
         5 father_1       9   10   11   13   12   14
        10 bbb           13   14   15   17   16   18The last part, starting with UNION, gets the "header " row.
The first part, before UNION, is exactly what I posted yesterday.
WITH     got_tree   AS
     SELECT     id
     ,     name
     ,     CONNECT_BY_ROOT id     AS root_id
     ,     CONNECT_BY_ROOT name     AS root_name
     ,     LEVEL               AS level_num
     ,     CASE
               WHEN id = PRIOR mother_id
               THEN 'M'
               ELSE 'F'
          END               AS mf_flag
     ,     SYS_CONNECT_BY_PATH ( CASE
                           WHEN id = PRIOR mother_id
                           THEN 'M'
                           ELSE     'F'
                          END
                        )     AS gender_path
     FROM     person     p
     CONNECT BY     (     id     = PRIOR mother_id
                   OR     id     = PRIOR father_id
          AND     LEVEL     <= :level_cnt
,     got_sa_num     AS
     SELECT     t.*
     ,     ROW_NUMBER () OVER ( PARTITION BY root_id
                               ORDER BY         level_num
                         ,                mf_flag     DESC
                         ,             gender_path     DESC
                       )      AS sa_num
     FROM     got_tree        t
SELECT     root_id
,     root_name
,     REPLACE ( SYS_CONNECT_BY_PATH ( TO_CHAR (id, '9999')
          )     AS ancestors
FROM     got_sa_num
WHERE     sa_num     = POWER (2, :level_cnt) - 1
START WITH     sa_num     = 2
CONNECT BY     sa_num     = PRIOR sa_num + 1
     AND     root_id     = PRIOR root_id
    UNION
SELECT      NULL          AS root_id
,     NULL          AS root_name
,     REPLACE ( SYS_CONNECT_BY_PATH ( LPAD ( SUBSTR ( REPLACE ( gender_path
                                    , 2
                              , 5     -- 1 + number of '9's in TO_CHAR, 13 lines up
          )     AS ancestors
FROM     got_sa_num
WHERE     sa_num     = POWER (2, :level_cnt) - 1
START WITH     sa_num     = 2
CONNECT BY     sa_num     = PRIOR sa_num + 1
     AND     root_id     = PRIOR root_id
ORDER BY root_id     NULLS FIRST
;

Hierarchical Queries help me

hi,
What is the usage of 'Connect' , 'Prior' in hierarchical queries and
what does it mean??

http://download-uk.oracle.com/docs/cd/B19306_01/server.102/b14200/queries003.htm
Cheers
Sarma.

Hierarchical queries no longer working with 11gr2

Hi, next hierarchical query:
SELECT REF(f) FROM STORE1.FOLDER_TABLE f
WHERE LEVEL > 1 START WITH REF(f) = ? CONNECT BY PRIOR f.key$ = f.header$.parentKey
used to work fine with 11gr1, while with 11gr2 it raises ORA-00904: "F": invalid identifier.
This seems to occur for all hierarchical queries.

Here's a simple script for SQL developer:
CREATE OR REPLACE TYPE Test_type AS OBJECT (
key RAW(16),
parentKey RAW(16))
CREATE TABLE Test_table OF Test_type (
key DEFAULT SYS_GUID() PRIMARY KEY)
OBJECT IDENTIFIER IS PRIMARY KEY;
SELECT REF(f) FROM Test_table f
START WITH f.parentKey IS NULL
CONNECT BY PRIOR f.key = f.parentKey
Returning ref(f) raises ORA-00904: "F": invalid identifier.
Returning f.key succeeds.

Hierarchical Queries / Rowsets via JDBC?

I have an application that requires data from an SQL database to be displayed in a hierarchical directed graph (i.e. a JTree). Is it possible to construct Hierarchical Queries that return Hierarchical Rowsets in servlets via JDBC? This would be akin to using a Microsoft ADODB Hierarchical Recordset object, but I need a Java solution that's portable and isn't dependent on proprietary extensions in a server (i.e. iPlanet, etc.).
Any help would be appreciated.
Thanks.

Is it possible to construct Hierarchical Queries that
return Hierarchical Rowsetsin servlets via JDBC?No. JDBC only speaks in flat rows and columns.
What type of hierarchical rowset are you referring to? Is it multidimensional data? For example MSSqlServer OLAP with row, column and measure dimensions? Or, is it simply grouped data for which you are both summarizing and displaying detail?

Self join vs Hierarchical Queries

Hi,
please tel me which one 1 should use ?
i have to get simple manger's employ list, should i use self join or hierarchical queries(CONNECT BY and prior)?
yours sincerely

Hi,
944768 wrote:
Hi,
please tel me which one 1 should use ?
i have to get simple manger's employ list, should i use self join or hierarchical queries(CONNECT BY and prior)?It depends on your data and your requirements.
Whenever you have a question, please post a little sample data (CREATE TABLE and INSERT statements) for all the tables involved, so the people who want to help you can re-create the problem and test their ideas. Also post the results you want from that data, and an explanation of how you get those results from that data.
Explain, using specific examples, how you get those results from that data.
If you can show what the problem is using commonly available tables (suc as scott.emp, which has a 4-level hierarchy) then you don't have ot post any sample data, just the results and the explanation.
Always say what version of Oracle you're using (e.g. 11.2.0.2.0).
See the forum FAQ {message:id=9360002}
If your hierarchy consists only of 2 levels, then a self-join will probably be more efficient, simpler to code, and easier to maintain.
If you don't know how many levels are in the hierarchy, then self-join isn't an option. Use CONNECT BY or, if you have Oracle 11.2, a recursive WITH clause.
If you have a fixed number of levels (or an upper bound) greater than 2, then CONNECT BY (or a recursive WITH clause) will probably be best.

Hierarchal queries and USER_DEPENDENCIES

I'm new to Hierarchal queries; I have a Hierarchal query on USER_DEPENDENCIES that does not work and gives an error
'TSPKG_ES' is a package the works; I'm trying to do query to see what is calls
SELECT LEVEL, D.*
FROM DBA_DEPENDENCIES D
START WITH D.name = 'TSPKG_ES'
CONNECT BY PRIOR D.name = D.referenced_name
Result = ORA-00600: internal error code, arguments: [sorsikbeg_1], [5], [0] ,[],[],[],[],[]
SELECT LEVEL, D.*
FROM DBA_DEPENDENCIES D
START WITH D.name = 'DATATESTSPKG_ES'
CONNECT BY PRIOR D.referenced_name = D.name
Result = ORA=0143: CONNECT BY loop in user data

OS = Microsoft Windows XP V2002 Service Pack 3
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - 64bi
PL/SQL Release 10.2.0.4.0 - Production
"CORE 10.2.0.4.0 Production"
TNS for 64-bit Windows: Version 10.2.0.4.0 - Production
NLSRTL Version 10.2.0.4.0 - Production

SELECT, hierarchical queries and JOIN

Hi everyone,
I have a small SELECT statement but I can't find an easy solution.
Look at this situation:
drop table departments;
CREATE TABLE departments
dpt_id NUMBER(10) UNIQUE,
dpt_name VARCHAR2(100),
dpt_parent_id NUMBER(10)
TRUNCATE table departments;
INSERT INTO departments VALUES(1, 'Company', null);
INSERT INTO departments VALUES(2, 'HR', 1);
INSERT INTO departments VALUES(3, 'SALES', 1);
INSERT INTO departments VALUES(4, 'IT', 1);
INSERT INTO departments VALUES(222, 'Helpdesk', 4);
INSERT INTO departments VALUES(223, 'French Speaking', 222);
INSERT INTO departments VALUES(224, 'Another level', 223);
INSERT INTO departments VALUES(5, 'LEGAL', 1);
INSERT INTO departments VALUES(66, 'Recruitment', 2);
INSERT INTO departments VALUES(33, 'Logistics', 2);
INSERT INTO departments VALUES(39, 'Fleet management', 33);
INSERT INTO departments VALUES(31, 'Local Sales', 3);
INSERT INTO departments VALUES(60, 'European Sales', 3);
INSERT INTO departments VALUES(61, 'Germany', 60);
INSERT INTO departments VALUES(62, 'France', 60);
INSERT INTO departments VALUES(620, 'Paris', 62);
INSERT INTO departments VALUES(621, 'Marseilles', 62);
INSERT INTO departments VALUES(38, 'American Sales', 3);
INSERT INTO departments VALUES(34, 'Asian Sales', 3);
CREATE table persons
person_id NUMBER(10) UNIQUE,
person_name VARCHAR2(100),
person_dpt_id NUMBER(10)
truncate table persons;
INSERT INTO persons VALUES(1, 'Jim', 2);
INSERT INTO persons VALUES(2, 'Jack', 621);
INSERT INTO persons VALUES(3, 'John', 620);
INSERT INTO persons VALUES(4, 'John', 224);
INSERT INTO persons VALUES(5, 'Fred', 61);It's a simple hierachy like the one we can find in HR schema. The link between an department and its parent is with parent id. THe following statement build the whole tree:
SELECT dpt_id, level, LPAD(' ', LEVEL-1)|| dpt_name
FROM departments
START WITH dpt_parent_id IS NULL
CONNECT BY dpt_parent_id = PRIOR dpt_id;As you can see in the script above, I have a few people assigned to these departments. It's also a classic situtation...
I would like to have something like this:
WITH temp AS
SELECT dpt_id, dpt_name, SYS_CONNECT_BY_PATH(dpt_name, '#') as full_path
    FROM departments
   START WITH dpt_parent_id IS NULL
CONNECT BY dpt_parent_id = PRIOR dpt_id
SELECT p.person_name, d.dpt_name, --d.full_path,
       regexp_substr(d.full_path, '[^#]+', 1, 2, 'i') as t1,
       regexp_substr(d.full_path, '[^#]+', 1, 3, 'i') as t2,
       regexp_substr(d.full_path, '[^#]+', 1, 4, 'i') as t3,
       regexp_substr(d.full_path, '[^#]+', 1, 5, 'i') as t4
FROM persons p
JOIN temp d ON d.dpt_id = p.person_dpt_id;This is the exact output I want, but I wonder... Is it possible to do it without the factored sub-query? It's nice and works fine but I had to precompute the whole path to split it again. I mean, this should be possible in one step. Any suggestion?
I'm using Oracle 10g
Thanks,

Hi,
user13117585 wrote:
... But sometimes, I just find the statements difficult for what they do. For example, my previous one. I have a person, and I want to see his department and the path in the tree.Actually, you want more than that; you want to parse the path, and display each #-delimited part in a separate column. If you didn't want that, then you could do away with the 4 REGEXP_SUBSTR calls, like this:
WITH temp AS
SELECT dpt_id, dpt_name
,       SUBSTR ( REPLACE ( SYS_CONNECT_BY_PATH ( RPAD (dpt_name, 15)     -- Using 15 just for demo
           , 16
           )     as full_path
    FROM departments
   START WITH dpt_parent_id IS NULL
CONNECT BY dpt_parent_id = PRIOR dpt_id
SELECT p.person_name, d.dpt_name, d.full_path
FROM persons p
JOIN temp d ON d.dpt_id = p.person_dpt_id;Output:
PERSON_N DPT_NAME      FULL_PATH
Jim      HR            HR
Fred     Germany       SALES          European Sales Germany
John     Paris         SALES          European Sales France         Paris
Jack     Marseilles    SALES          European Sales France         Marseilles
John     Another level IT             Helpdesk       French SpeakingAnother levelAs you can see, full_path is one giant column, but it's formatted to look like 4 separate columns, forresponding to your original t1, t2, t3 and t4. I limited the output to 15 characters, just for debugging and posting purposes. You can use any number of characters you like.
It's too complex for this simple thing.It would be nice if there was something simpler that did exactly what you wanted, but I'm not sure it's reasonable to expect it in every case. I asked a lot of questions in my first message, but I'm not sure you've tried to answer any of them, so I'm not sure why you're unhappy with the query you posted. I can think of lots of ways to change the query, but I have no way of telling if you would like them any better than what you already have.
And hopefully, I know where to start in the hierarchy and I know where to stop. If I had to show all the levels and have one column by level dynamically, I'd be stuck. Sorry, I don't understand this part.
Are you saying that it seems inefficient to generate the entire tree, when perhaps few of the nodes will have have matches in the persons table? If so, you can invert the whole query. Instead of doing the CONNECT BY first and then joining, do the join first and then the CONNECT BY. Instead of doing a top-down CONNECT BY, where you start with the parentless nodes (whether or not you'll ultimately need them) and then find their descendants, do a bottom-up CONNECT BY, where you start with the nodes you know you'll need, and then find their ancestors.
I just find it difficult for such a simple need. Again, there are lots of things that could be done. If you won't say what you want, that makes it hard for me to tell you how to get it. All that I've picked up for sure is that you don't like doing a sub-query. That's unfortunate, because sub-queries are so basic. They have very important been since Oracle 8.1, and they don't seem to be going away. Quite the opposite, in fact. You need sub-queries for all kinds of things, not just CONNECT BY. To give just a couple of examples, they're the only thing that make analytic functions really useful, and they simplfy chasm traps (basically, multiple 1-to-many relationships on the same table) considerably. I'm sorry if you don't lke sub-queries, but I don't see how you can work in this field and not use them.
Edited by: Frank Kulash on Nov 15, 2011 3:18 PM
Revised query

Top Link 10.1.3 is it possible to define hierarchical queries in jdev

Is it possible to define a hierarchical query in a named query in the mapping workbench (jdev 10.1.3) or we have to do it dynamically with the java api ?
Any example is welcome.
Best regards

i didn't find the method that return the readallquery object i need from my named query
i'm doing something wrong ?
public class FlexTreeSetHierarchicalClause
public static void setHierarchicalClause(oracle.toplink.descriptors.ClassDescriptor descriptor)
    // instantiante expression builder to build the expressions
    ExpressionBuilder expressionBuilder = new ExpressionBuilder();
    // Specifies a START WITH expression
    Expression startExpr = expressionBuilder.get("Flextree").get("flexitemI").equal(expressionBuilder.get("Flextreenode").get("flexitemparentI"));
    // Specifies a CONNECT BY expression
    Expression connectBy = expressionBuilder.get("Flextreenode").get("flexitemchildI").equal(expressionBuilder.get("Flextreenode").get("flexitemparentI"));
    //Specifies an ORDER SIBLINGS BY vector
    Vector order = new Vector();
    order.addElement(expressionBuilder.get("Flextreenode").get("flexitemchildI").get("flextext").get("label"));
    // get the readallobject to add the hierarchical clause
    DatabaseQuery dq = descriptor.getQueryManager().getQuery("getFlexTreeByName");
    ReadAllQuery raq = descriptor.getQueryManager().getReadAllQuery();
    // add the hierarchical clause
    raq.setHierarchicalQueryClause(startExpr, connectBy, order);

Problem with hierarchical queries

Hi,
I implemented (thx to jarola) sucessfully a report using a hierarchical view.
It is nearly the same as in Tyler Muths blog: -http://tylermuth.wordpress.com/2009/02/26/hierarchical-query-to-unordered-list/-
-select (level*20)-20 the_level ,(level*10)-10 the_level2, machine_no
- parent_id, item_id,description
-from machine_list
-where machine_no = 101
-and project_id = 122
- connect by prior item_id = parent_id
- start with parent_id is null
This works good, as long as there is just one project/machine in the table.
If I have 2 machines in the table, it does not work in the report anymore, all entries are shown, although I check for the right machine_number and project_number in the query :-(
Any suggestions?
BR
Lena

Hi Joel
Just for reference, to write the code tags in posts without a space, you can use noformat tags on either side like this (let's see if I can manage this!)...
<pre>{noformat }{noformat }</pre>
Edited by: Munky on Apr 27, 2011 10:03 AM
Edited by: Munky on Apr 27, 2011 10:03 AM
Edited by: Munky on Apr 27, 2011 10:03 AM
Edited by: Munky on Apr 27, 2011 10:04 AM
Edited by: Munky on Apr 27, 2011 10:04 AM
Edited by: Munky on Apr 27, 2011 10:05 AM - sod it! had to add spaces to the noformat tags!

Hierarchical queries and referential constraints

Experts,
What is the rationale for having a referential integrity constraint on a table that has rows bound by hierarchical relationships ?? The hierarchical relationship ensures that each row is referencing another in the same table via the PRIOR clause. If I add a foreign key relationship to these columns , does it make any sense ??/
-Rekha

but you can use a hierarchical query to display constraints in a tree !
SQL> create table t1 (a number primary key);
Table created.
SQL> create table t2 (b number primary key, a number references t1(a));
Table created.
SQL> create table t3 (c number primary key, b number references t2(b));
Table created.
SQL> create table t4 (d number primary key, a number references t1(a));
Table created.
SQL> create table t5 (e number primary key, f number references t5(e));
Table created.
SQL> select sys_connect_by_path(p.table_name,':')||':'||r.table_name p
from all_constraints p
join all_constraints r
on (p.constraint_name = r.r_constraint_name and p.owner = r.r_owner)
where CONNECT_BY_ISLEAF = 1
connect by nocycle prior r.table_name = p.table_name
start with p.table_name not in (select r.table_name from all_constraints p join
all_constraints r on (p.constraint_name = r.r_constraint_name and
p.owner = r.r_owner and p.table_name != r.table_name) );
:DEPT:EMP
:T1:T4
:T1:T2:T3
:T5:T5

Hierarchical Queries - Challenging

Similar Messages

Maybe you are looking for