Check on spatial column

I have a table Entity2 with two spatial attributes (ghost and geometry) and I want add a topological constraints between them. Why does the following sql string do an error?
ALTER TABLE Entity2 ADD CHECK (SDO_GEOM.RELATE(ghost,'inside+coveredby',geometry,0.005));

This is an area I have been looking into a lot recently and am putting the final touches together on a "white paper" on this topic to be hosted on my website.
The Oracle documentation says:
Conditions of check constraints cannot contain the following constructs:
1 Subqueries and scalar subquery expressions
2 Calls to the functions that are not deterministic (CURRENT_DATE, CURRENT_TIMESTAMP, DBTIMEZONE, LOCALTIMESTAMP, SESSIONTIMEZONE, SYSDATE, SYSTIMESTAMP, UID, USER, and USERENV)
3 Calls to user-defined functions
4 Dereferencing of REF columns (for example, using the DEREF function)
5 Nested table columns or attributes
6 The pseudocolumns CURRVAL, NEXTVAL, LEVEL, or ROWNUM
7 Date constants that are not fully specified
Now 1 is allowed in the SQL92 standard (ie CHECK constraints with SubQueries or via ASSERTIONS) but Oracle does not support them.
2 is a problem because sdo_geom.relate is not deterministic as can be seen from:
SELECT object_name,
       procedure_name,
      deterministic
FROM all_procedures
WHERE owner = 'MDSYS'
   AND object_name = 'SDO_GEOM'
   AND procedure_name = 'RELATE'3 means you can't create your own function around this whose return type would be deterministic. Even if SDO_GEOM.RELATE was deterministic you can't use it because it is still classed as a user-defined (MDSYS user) function (try it with one of the SDO_GEOM deterministic functions etc SDO_MBR() IS NOT NULL to see that this is the case).
The others don't affect this discussion.
The only way that I can see that you can do this is to create a trigger.
create trigger entity2_bi before insert
on entity2
As
begin
IF sdo_geom.relate(:new.ghost,'inside+coveredby',:new.geometry,0.005) <. 'TRUE' THEN
raise_application_error(020000,'GHOST is not inside or covered by GEOMETRY',true);
END IF;
End;
Hope this helps
regards
S

Similar Messages

Is it mandatory to populate USER_SDO_GEOM_METADATA for each spatial column ?

Hello,
I would like to know if we HAVE to populate the metadata table for a geometry column that will never be indexed.
(I know that in the documentation it is said "For each spatial column,
you must insert an appropriate row into the USER_SDO_GEOM_METADATA view.")
Thanks
Hugues

Hi Hugues,
There are many spatial operations that make use of the user_sdo_geom_metadata table, not all of them index related (operators).
If your data hasn't been migrated to 8.1.6 format then not much will work if user_sdo_geom_metadata is not populated.
If your data is migrated:
You won't be able to validate your data (we check the data to the bounds of the coordinate system).
You won't be able to use linear referencing (also known as dynamic segmentation).
You won't be able to use any spatial operators (which require a spatial index).
You won't be able to use coordinate systems (or least some parts of coordinate systems).
The above might only be a partial list.
hope this helps, and regards,
dan

How to avoid to check if a column value is NULL?

Hi, I'm a newbee in Oracle.
I have a procedure like this:
create or replace
PROCEDURE get_employee
     v_first_name IN VARCHAR2 DEFAULT NULL ,
     v_middle_name IN VARCHAR2 DEFAULT NULL ,
     v_last_name IN VARCHAR2 DEFAULT NULL ,
To select rows with matching multiple column values, I can simply do this:
SELECT *
FROM employee
WHERE first_name = v_first_name
AND middle_name = v_middle_name
AND last_name = v_last_name
The problem is v_middle_name can be NULL. This means,
I need check if v_middle_name is NULL, and if it is, I need use "IS NULL" instead, like this:
SELECT *
FROM employee
WHERE first_name = v_first_name
AND middle_name IS NULL
AND last_name = v_last_name
It seems very cumbersome to do a check for each column that can be null.
Is there a way that I do not need to do a check for every column?
or is it better to avoid having NULL values in those columns (and replace them with, say a space) ?
Thanks in advance.
Simon

Normally, you would do something like
SELECT *
FROM employee
WHERE first_name = NVL( v_first_name, first_name )
   AND middle_name = NVL(v_middle_name, middle_name )
   AND last_name = NVL(v_last_name, last_name )Of course, if you can ensure that NULL data is not allowed (without creating phony non-NULL data), that is a good thing. In most systems, for example, it is probably reasonable to require a non-NULL first and last name. But you almost certainly cannot require a middle name.
Justin

Using '=' to check for NULL column value

We are currently encountering an issue where 10.1.0.5 is happily accepting a check on a column for a null value with what I have always known to be an invalid syntax.
For instance:
select col1 from my table where col1=null
Although this is wrong and we are working on locating the queries formatted as such, what I am wondering is why is it working on the 10.1.0.5 version but we cannot get the same query to execute on 8i, 9i, 10.2.0...? Is there a flag that can be enabled to allow this compare?

I am familiar with the correct syntax to be using. What I am looking for is a possible explanation as to why the query below returns the correct results on 10.1.0.5 and returns nothing in all other versions. I would expect the same behavior across all instances. See the two lines in bold "AREA=NULL" and "SUBAREA=NULL"
SELECT DISTINCT
UNIQID,
MGRCOMMENT,
REQ.CREATED_BY CREATED,
C2.CNAME DEVELOPER,
C5.CNAME PRODMGR,
C3.CNAME ASSIGNED,
C4.CONTACTID CURRENTUSER,
ID,
STATUSNAME,
REQ.STATUS STATUSID,
PRIORITYNAME,
TITLE,
MAS1.MODULENAME,
MAS1.AREANAME,
MAS1.SUBAREANAME,
MIN1.MODULE,
MIN2.AREA,
MIN3.SUBAREA
FROM REQ,
PRIORITY,
STATUS,
(SELECT FIRST_VALUE(M1.MODULENAME)
OVER (PARTITION BY REQUNIQID ORDER BY MODULENAME ROWS UNBOUNDED PRECEDING) MODULENAME, FIRST_VALUE(A1.AREANAME)
OVER (PARTITION BY REQUNIQID ORDER BY AREA ROWS UNBOUNDED PRECEDING) AREANAME, FIRST_VALUE(SA1.SUBAREANAME)
OVER (PARTITION BY REQUNIQID ORDER BY SUBAREANAME ROWS UNBOUNDED PRECEDING) SUBAREANAME, REQUNIQID
FROM REQMOD, MOD M1, ARA A1, SBARA SA1
WHERE M1.MODULEID = REQMOD.MODULE
AND A1.AREAID = REQMOD.AREA
AND SA1.SUBAREAID = REQMOD.SUBAREA) MAS1,
(SELECT FIRST_VALUE(C2.CNAME) OVER (PARTITION BY REQUNIQID ORDER BY CNAME ROWS UNBOUNDED PRECEDING) CNAME, REQUNIQID
FROM REQ_ASSIGN, CONTACT C2
WHERE C2.CONTACTID = REQ_ASSIGN.CONTACTID
AND ASSIGNMENT_TYPE=2) C2,
(SELECT FIRST_VALUE(C5.CNAME) OVER (PARTITION BY REQUNIQID ORDER BY CNAME ROWS UNBOUNDED PRECEDING) CNAME, REQUNIQID
FROM REQ_ASSIGN, CONTACT C5
WHERE C5.CONTACTID = REQ_ASSIGN.CONTACTID
AND ASSIGNMENT_TYPE=1) C5,
(SELECT FIRST_VALUE(C3.CNAME)
OVER (PARTITION BY REQUNIQID ORDER BY CNAME ROWS UNBOUNDED PRECEDING) CNAME, REQUNIQID
FROM REQ_ASSIGN, CONTACT C3
WHERE C3.CONTACTID = REQ_ASSIGN.CONTACTID
AND ASSIGNMENT_TYPE<>2
AND ASSIGNMENT_TYPE<>1) C3,
(SELECT DISTINCT REQUNIQID, CONTACTID FROM REQ_ASSIGN WHERE CONTACTID=5) C4,
(SELECT DISTINCT REQUNIQID, MODULE FROM REQMOD WHERE MODULE=1) MIN1,
*(SELECT DISTINCT REQUNIQID, AREA FROM REQMOD WHERE AREA=NULL) MIN2,*
*(SELECT DISTINCT REQUNIQID, SUBAREA FROM REQMOD WHERE SUBAREA=NULL) MIN3,*
REQCOMMENTS
WHERE PRIORITY.PRIORITYID = REQ.PRIORITY
AND STATUS.STATUSID = REQ.STATUS
AND MAS1.REQUNIQID = REQ.UNIQID
AND C2.REQUNIQID = REQ.UNIQID
AND C5.REQUNIQID = REQ.UNIQID
AND C3.REQUNIQID = REQ.UNIQID
AND C4.REQUNIQID = REQ.UNIQID
AND MIN1.REQUNIQID = REQ.UNIQID
AND MIN2.REQUNIQID = REQ.UNIQID
AND MIN3.REQUNIQID = REQ.UNIQID
AND REQCOMMENTS.REQUNIQID = REQ.UNIQID
AND REQ.PROJECTID = 3;

Check box als column in a standard table, how to get the selected row

Dear experts,
I habe standard tablt with check box as column. Now I want to get the current selected row structure and do some changes. How could I solve this problem? till now I just know to get the structure via lead selection.
lo_node->get_element().
lo_element = lo_node->get_static_attributes ( static_attributes = ls_row).
How could I get the element through check-box in stead of lead selection. Many thanks!

check this code
To get the selected row number
data: lr_element type ref to if_wd_context_element,
lv_index type i.
lr_element = wdevent->get_context_element( name = 'CONTEXT_ELEMENT' ).
lv_index = lr_element->get_index( ).
Thanks
Bala Duvvuri

Replicate Table with spatial columns

I'd like to know how to
replicate (updatable snapshot) a table
with spatial columns.
when i try to create a master repobject
i've got ORA-23318 error.
What can i do?
null

Hi Vincent,
Oracle does not support replication of object in 8.1.6 or 8.1.7. I believe replication support is expected for the version after.
Sorry,
Dan

Problem by adding one spatial column to the version-enabled table

Hello, I'm trying to modify one version-enabled table T1 like this:
--table definition
create table T1
(ID NUMBER NOT NULL PRIMARY KEY,
NAME VARCHAR2 (256),
FUNCTION VARCHAR2 (256));
--enable versioning
EXECUTE DBMS_WM.EnableVersioning('T1','VIEW_WO_OVERWRITE');
I'd like to add one spatial column to this table by:
-- modify metada view for spatial indexing
INSERT INTO USER_SDO_GEOM_METADATA (TABLE_NAME, COLUMN_NAME, DIMINFO, SRID)
VALUES ('T1_LT', 'ANCHOR',
MDSYS.SDO_DIM_ARRAY
(MDSYS.SDO_DIM_ELEMENT('X', 0.000, 100000.000, 0.0005),
MDSYS.SDO_DIM_ELEMENT('Y', 0.000, 100000.000, 0.0005),
MDSYS.SDO_DIM_ELEMENT('Z', -100, 1000, 0.0005)) , 81989002);
-- table modification - add a column and create a spatial index
EXECUTE DBMS_WM.BeginDDL('T1');
ALTER TABLE T1_LTS ADD ("ANCHOR" MDSYS.SDO_GEOMETRY);
CREATE INDEX T1_SPX on T1_LTS(ANCHOR) INDEXTYPE is MDSYS.SPATIAL_INDEX;
EXECUTE DBMS_WM.CommitDDL('T1');
By finishing of the DDL operation with EXECUTE DBMS_WM.CommitDDL('T1') I get an error message:
"SQL> EXECUTE DBMS_WM.CommitDDL('T1');
BEGIN DBMS_WM.CommitDDL('T1'); END;
ERROR at line 1:
ORA-20171: WM error: 'CREATE TABLE' and 'CREATE SEQUENCE' privileges needed.
ORA-06512: at "SYS.WM_ERROR", line 342
ORA-06512: at "SYS.WM_ERROR", line 359
ORA-06512: at "SYS.LTUTIL", line 8016
ORA-06512: at "SYS.LT", line 11925
ORA-06512: at line 1
What is wrong here? The Oracle 10g DB is installed on Windows 2003 Server, OWM_VERSION - 10.2.0.1.0.
Regards,
Viktor

Hi,
You need to explicitly grant the create table and create sequence privileges to the user owning the table. It is not enough for these to be granted by using a role. This restriction is documented in the user guide.
Also, you should add the entry in the user_sdo_geom_metadata view on the t1_lts table, which is the table to which you are actually adding the geometry column, after calling beginDDL. CommitDDL will make the necessary changes, which in this case would be to add an entry for both t1 and t1_lt.
Regards,
Ben

Integrity check fails on column key

Dear DBA's,
I am getting error while creating encrypted tablespace "ORA-S8342: Integrity check fails on column key"
actually i am trying to work on Wallet,
I created wallet key without auto login option initially and i created the tablespaces, user and tables associated to the tablespace.
after that i dropped the tablespace, user and tables as well as wallet key using Metalink ID 757936.1.
after that i created wallet key using orapki option after I tried to create the tablespace it is giving above error.
kindly help me how to resolve this issue.
I have one doubt that we can create encrypted tablespace only ones for a database instance?
to create one more encrypted tablespace what will be the procedure to remove the previous encrypted tablespace?
Best Regards,
SG

after that i created wallet key using orapki option after I tried to create the tablespace it is giving above error.Please log a SR for this issue.
I have one doubt that we can create encrypted tablespace only ones for a database instance?You can create more than one.
to create one more encrypted tablespace what will be the procedure to remove the previous encrypted tablespace?Drop the encrypted tablespace like dropping any other tablespace -- Use "DROP TABLESPACE <tablespace name> INCLUDING CONTENTS AND DATAFILES;" command.
See this link in addition to Oracle documentation -- http://www.oracle-base.com/articles/11g/TablespaceEncryption_11gR1.php
Thanks,
Hussein

Putting a check box above column for selection in WD ALV

Hi,
Can anyone provide me a code snippet and help me how I can
provide a check box above column for selection in WD ALV.
Best Regards
Sid

Hi Sid,
As far as I know we cannot put checkboxes directly above the Column headers.
You can achieve this functionality in a round above method
You can create a group of checkboxes above your ALV table with names of columns to be selected beside each check box and proceed with your further processing.
Also If you are just looking at which columns you need to select to be shown on the
ALV output
data: lr_standard_functions type ref to if_salv_wd_std_functions.
lr_standard_functions ?= wd_this->r_table.
lr_standard_functions->set_column_selection_allowed( ' ' ).
Hope this is of some help.
Regards,
Ismail.
Edited by: Ismail Basha on Jan 23, 2008 1:47 PM
Edited by: Ismail Basha on Jan 24, 2008 3:23 AM

Performance problem on view with spatial column - resolved

I have had a problem with queries on a view that had a spatial column, where the view did not belong to the logged in user. Where my spatial window was retrieved by a sub-query, the spatial scan did a full table scan instead of using the spatial index.
I have found that the problem can be resolved by granting MERGE VIEW on the view to the querying user.
The view can be as simple as SELECT * FROM table.
The badly performing query could be as simple as
select id from T1.tstview
where SDO_RELATE(coordinates,
(SELECT coordinates FROM T1.tstWINDOW WHERE ID = '1')
,'mask=INSIDE+COVEREDBY querytype=WINDOW') = 'TRUE' ;
I think this is a bug, and have raised an SR - MERGE VIEW is supposed to override issues with the "security intent" of a view.
The workaround is simple enough once you're aware of it and I thought it was worth passing on.

Thanks for sharing this workaround!
Which ORACLE version did you test ?

Column alias for spatial column within cursor loop using dynamic SQL

The following PL/SQL is trying to generate an error report for records or objects which are 3 dimensional or above. I have no issue execute one statement in SQLPLUS but I need to use the column alias for the spatial column. But, it is a different story using PL/SQL and dynamic SQL Any help will be great because I've been working on this for than 8 hours but with no luck! Thanks.
Here is the error I'm getting,
stmt := 'select p.column_name.get_gtype(), id from '|| table_name p ' where p.column_name.get_gtype() > 2 ';
ERROR at line 15:
ORA-06550: line 15, column 79:
PLS-00103: Encountered the symbol "P" when expecting one of the following:
. ( * @ % & = - + ; < / > at in is mod remainder not rem
<an exponent (**)> <> or != or ~= >= <= <> and or like LIKE2_
LIKE4_ LIKEC_ between || member SUBMULTISET_
The symbol "* was inserted before "P" to continue.
and my PL/SQL is,
set serveroutput on size 100000 feedback off
declare
rs integer;
rs1 integer;
cur integer;
rp integer;
trs integer;
n integer;
un varchar2(30);
stmt varchar2(200);
begin
dbms_output.put_line(rpad('Table Name',40)||' Dimension');
dbms_output.put_line(rpad('-',53,'-'));
cur:= dbms_sql.open_cursor;
for t in (select column_name,table_name from user_sdo_geom_metadata where regexp_like(table_name, '[^[A-B]_[AB]$'))
loop
stmt := 'select p.column_name.get_gtype(), id from '|| table_name p ' where p.column_name.get_gtype() > 2 ';
dbms_sql.parse(cur, stmt, dbms_sql.native);
dbms_sql.define_column(cur, 1, rs);
dbms_sql.define_column(cur, 2, rs1);
rp:= dbms_sql.execute(cur);
n:=dbms_sql.fetch_rows(cur);
dbms_sql.column_value(cur, 1, rs);
dbms_sql.column_value(cur, 2, rs1);
dbms_output.put_line(rpad(t.table_name,38,'.')||rpad(rs,15)||rpad(rs1,15));
end loop;
dbms_sql.close_cursor(cur);
dbms_output.put_line(rpad('-',53,'-'));
end;
set serveroutput off feedback on feedback 6

The following PL/SQL is trying to generate an error report for records or objects which are 3 dimensional or above. I have no issue execute one statement in SQLPLUS but I need to use the column alias for the spatial column. But, it is a different story using PL/SQL and dynamic SQL Any help will be great because I've been working on this for than 8 hours but with no luck! Thanks.
Here is the error I'm getting,
stmt := 'select p.column_name.get_gtype(), id from '|| table_name p ' where p.column_name.get_gtype() > 2 ';
ERROR at line 15:
ORA-06550: line 15, column 79:
PLS-00103: Encountered the symbol "P" when expecting one of the following:
. ( * @ % & = - + ; < / > at in is mod remainder not rem
<an exponent (**)> <> or != or ~= >= <= <> and or like LIKE2_
LIKE4_ LIKEC_ between || member SUBMULTISET_
The symbol "* was inserted before "P" to continue.
and my PL/SQL is,
set serveroutput on size 100000 feedback off
declare
rs integer;
rs1 integer;
cur integer;
rp integer;
trs integer;
n integer;
un varchar2(30);
stmt varchar2(200);
begin
dbms_output.put_line(rpad('Table Name',40)||' Dimension');
dbms_output.put_line(rpad('-',53,'-'));
cur:= dbms_sql.open_cursor;
for t in (select column_name,table_name from user_sdo_geom_metadata where regexp_like(table_name, '[^[A-B]_[AB]$'))
loop
stmt := 'select p.column_name.get_gtype(), id from '|| table_name p ' where p.column_name.get_gtype() > 2 ';
dbms_sql.parse(cur, stmt, dbms_sql.native);
dbms_sql.define_column(cur, 1, rs);
dbms_sql.define_column(cur, 2, rs1);
rp:= dbms_sql.execute(cur);
n:=dbms_sql.fetch_rows(cur);
dbms_sql.column_value(cur, 1, rs);
dbms_sql.column_value(cur, 2, rs1);
dbms_output.put_line(rpad(t.table_name,38,'.')||rpad(rs,15)||rpad(rs1,15));
end loop;
dbms_sql.close_cursor(cur);
dbms_output.put_line(rpad('-',53,'-'));
end;
set serveroutput off feedback on feedback 6

Af:table - Programmatic mandatory check on a column

I have a usecase wher in, i need to programatically apply a mandatory check on a column inside a table.
User can add multiple rows in a table and only on the click of a button(say "Save"), all the rows in the table should be validated and those with missing value for mandatory column fields should be marked in RED. Say if user, adds 4 rows consecutively and he doesnt provide value for mandatory column of first and last rows, then tht specific fields should be marked in red.
Any pointers?

Hi,
create a boolen variable in managed bean and bind with mandatory field.
<af:inputText>
required="#{EditDetails.valueReqXX}"
binding="#{EditDetails.reasonCommentXX}"
</af:inputText>managed bean code :
private Boolean valueReqXX;
private RichInputText reasonCommentXX;
    public void setValueReqFhlmc(Boolean valueReqXX) {
        this.valueReqXX = valueReqXX;
    public Boolean getValueReqXX() {
        return valueReqXX;
    public void setReasonCommentXX(RichInputText reasonCommentXX) {
        this.reasonCommentXX= reasonCommentXX;
    public RichInputText getReasonCommentXX() {
        return reasonCommentXX;
public String buttonAction()
                valueReqXX = true;//or false based on logic
                AdfFacesContext.getCurrentInstance().addPartialTarget(reasonCommentXX);
return null;
}~Abhijit
Edited by: Abhijit Dutta on Oct 20, 2011 6:19 PM
Edited by: Abhijit Dutta on Oct 20, 2011 7:02 PM

Specifying spatial column in USER_SDO_GEOM_METADATA with subtype

How do I specify the spatial column in USER_SDO_GEOM_METADATA when the
column is of a subtype? I have a table of type 'SPATIAL_OBJECT' with
subtypes for links and nodes, containing the SDO_GEOMETRY attributes.
TREAT does not appear to work.

"The problem I have is making the spatial columns of the subtypes visible in USER_SDO_GEOM_METADATA. Its also not a function - I have the geometry objects in the subtypes, I just need to make them visible to index them." - but this is exactly what is being shown in the example.
For instance, if I have a table called 'cola_markets_2' with a column called 'market', that is made up of a user-defined type called 'market_types' whereby 'shape' is of the type SDO_GEOMEMETY and all I want to do is insert the reference of the sub-type for into USER_SDO_GEOM_METADATA for indexing purposes I can simply use dot notation to do so:
INSERT INTO user_sdo_geom_metadata
(TABLE_NAME,
COLUMN_NAME,
DIMINFO,
SRID)
VALUES (
'cola_markets_2',
'market.shape',
SDO_DIM_ARRAY( -- 20X20 grid
SDO_DIM_ELEMENT('X', 0, 20, 0.005),
SDO_DIM_ELEMENT('Y', 0, 20, 0.005)
NULL -- SRID
Then, to index the table and sub-type I use the same dot notation:
CREATE INDEX cola_spatial_idx_2
ON cola_markets_2(market.shape)
INDEXTYPE IS MDSYS.SPATIAL_INDEX;
Have you tried this yet? This will not break anything and from what you say this is exactly what you are looking for. I guess I am not understanding why you would be looking at functions or member functions when you say you "have the geometry objects in the subtypes" and "just need to make them visible to index them". This is how to do it.
Does this make sense?
-Justin

Modify the Data Load wizard to remove the check for unique columns

Hi,
I'm trying to make a Data Loading Wizard into my application, which basically works nice except for one thing:
I want to remove the check for unique columns, and insert ALL data which the users wants to upload.
(Like the Data load utility under development).
The reason is, that the data I want to upload is an export from my bank statements (in XLS / CSV format)
And these exports don't have any primary key.
Using the Apex wizard created Data Upload pages, I will loose data in case I had two identical payments in a day.
Is there a way to do this, while keeping the pages created by the wizard to make Data Loading pages?
apex 4.2.1

I would suggest to load the data into a view and process the records into the real table(s) with instead-of-triggers. When you add a "1=2" where clause to the view, there never is any data and everything will be loaded.
Otherwise you have to have a "real" PK value, like a timestamp or a statement_nr+line_nr.

Oracle Spatial column in OWB

Does OWB 10g (10.1) supports Oracle Spatial Columns?
eg: SDO_GEOMETRY column
As far as i know timestamp datatype is not supported by OWB. Can anybody list out the unsupported datatypes by OWB?
Thanks in advance.
Sathish K

I am not sure if OWB 10g R1 supports Oracle Spatial Data Type. This spatial data type is a problem in many ETL tools.
I had come across a situation where it was needed to migrate spatial data from legacy databases. If OWB does not support then what can be done is to call a stored procedure in the mapping and pass the input parameters and let the procedure handle the inserts/updates/deletes, whatever you feel like.
This way you give all the control of data loading to the database, but this is the only way that I can think of.
Do let me know if you could find a smarter way of doing this.
Regards
-Arnab

Check on spatial column

Similar Messages

Maybe you are looking for