SpatialDB Advisor
There have been many, many articles written on the web describing how to access external data (data held in operating system files not under the control of Oracle and not stored within its own datafiles and tablespaces) from within Oracle such that one can query the data using SQL.
In short, Oracle allows a user to access external data by wrapping a table definition over the top of the external data files and using its SQL loader and DataPump technologies to convert the data “on the fly” as users query it. As the Oracle documentation notes, this is a read-only table whose metadata is stored in the database but whose data in stored outside the database.
The table “wrapper” is achieved by using the CREATE TABLE statement to define the columns of the table in the normal manner, but the appending to the definition its physical_properties, in particular, the ORGANIZATION EXTERNAL clause.
This clause is made up of two dependent sub-clauses, the external_table_clause and the external_data_properties that allows you to specify the properties so described. For example, the TYPE clause of external_table_clause allows you to indicate the driver to be used to access external table’s data. Oracle database provides two access drivers: ORACLE_LOADER and ORACLE_DATAPUMP. Whereas the ACCESS PARAMETERS clause of “external_data_properties” allows you to describe the physical layout of the file (what type of delimiter, number of fields etc).
The best way to describe the external data loading capability of Oracle is to show you how to use it to load a small sample of LiDAR data (after it has been post-processed).
Firstly, we need to place our data in an operating system directory which Oracle can access, (in this example I have placed a file called ground.csv in C:\Temp\Lidar), and then create an Oracle Directory object and grant the user that will create and load the data (here my standard “CODESYS” user), read access on that directory.
-- Create directory that points to actual data file.
CREATE OR REPLACE DIRECTORY EXTERNAL_DATA_DIRECTORY AS 'C:\temp\lidar';
Now we need to define our table object in Oracle. We also need to describe its external properties as part of the one definition.
DROP TABLE lidar;
Now we execute SQL SELECT statements against the table.
SELECT count(*)
FROM lidar;
I will now demonstrate that you cannot create spatial indexes on this table. Firstly, note that I have no SDO_GEOMETRY column in this table that I can index. However, I could try creating a function-based index (even though I doubt it would work because Oracle indexes normally contain ROWIDs that link the index to the actual row in the table holding the data. External tables do not have ROWIDs.)
It is worth noting the following from the Oracle documentation.
Restrictions on External Tables
External tables are subject to the following restrictions:
- An external table cannot be a temporary table.
- You cannot specify constraints on an external table.
- An external table cannot have object type, varray, or LONG columns.
(I will look at the last limitation at the end of this article.)
Let’s create the function based index in the common manner, including the metadata.
-- Create Function that returns a point geometry so we can create a function-based index
--
create or replace function observation(p_srid in number,
p_x in number,
p_y in number,
p_z in number )
return mdsys.sdo_geometry deterministic
Is
Begin
Return MDSYS.SDO_GEOMETRY(3001,p_srid,mdsys.sdo_point_type(p_x,p_y,p_z),NULL,NULL);
End observation;
/
show errors
Don’t you love the recommended Action!
OK, so our external data of LiDAR points can really on read via full-table scans as one cannot create an index on an externally organized table.
If we want to spatially index we have to bring the data inside the database which we will now do.
-- Rename our existing LiDAR table to LiDAR_EXT
DROP TABLE lidar_ext;
So, now we can query the data.
SELECT count(*)
FROM lidar l
WHERE SDO_ANYINTERACT(l.geom,sdo_geometry(2003,28355,null,sdo_elem_info_array(1,3,3),sdo_ordinate_array(478040,5228050,478060,5228070))) = 'TRUE'
AND l.geom.sdo_point.z BETWEEN 200 AND 600;
Data Other than Points
Our example is based on point data that is loaded into three numeric attributes rather than to an SDO_GEOMETRY object directly. As noted above:
- An external table cannot have object type, varray, or LONG columns.
Instead of modifying my existing example to try and define the SDO_GEOMETRY object directly on the X, Y and Z fields in the CSV, I thought I would attempt to load the linear data associated with my article on executing shp2sdo from within the database.
The shp2sdo utility creates a SQL Loader control file with the actual data stored in-line inside the control file or externally in a data file. This is controlled by the “-d” parameter. I re-ran the tool so that I got the shapefile’s data in the separate data file. I the opened the CTL file, grabbed the load parameters and constructed an appropriate CREATE TABLE statement from it inside Oracle database. Here is that statement:
CREATE OR REPLACE DIRECTORY TEMP_DATA_DIRECTORY AS 'C:\temp';
On execution this is what happened.
Error starting at line 4 in command:
CREATE TABLE GEODATA250K_TAS_ROADS
( GID NUMBER(38,0),
CLASS VARCHAR2(20 BYTE),
FEATWIDTH VARCHAR2(20 BYTE),
NRN VARCHAR2(20 BYTE),
PID VARCHAR2(20 BYTE),
FEATREL VARCHAR2(20 BYTE),
FORMATION VARCHAR2(20 BYTE),
ATTRREL VARCHAR2(20 BYTE),
RETIRED VARCHAR2(20 BYTE),
SRN VARCHAR2(20 BYTE),
SOURCE VARCHAR2(20 BYTE),
SHAPE_LENG VARCHAR2(20 BYTE),
NAME VARCHAR2(20 BYTE),
TEXTNOTE VARCHAR2(20 BYTE),
FEATTYPE VARCHAR2(20 BYTE),
PLANACC VARCHAR2(20 BYTE),
SYMBOL VARCHAR2(20 BYTE),
CREATED VARCHAR2(20 BYTE),
PEANO NUMBER,
GEOM MDSYS.SDO_GEOMETRY
)
ORGANIZATION EXTERNAL
(TYPE ORACLE_LOADER
DEFAULT DIRECTORY TEMP_DATA_DIRECTORY
ACCESS PARAMETERS (
CONTINUEIF NEXT(1:1) = '#'
FIELDS TERMINATED BY '|'
TRAILING NULLCOLS (
GID INTEGER EXTERNAL,
CLASS NULLIF CLASS = BLANKS,
FEATWIDTH NULLIF FEATWIDTH = BLANKS,
NRN NULLIF NRN = BLANKS,
PID NULLIF PID = BLANKS,
FEATREL NULLIF FEATREL = BLANKS,
FORMATION NULLIF FORMATION = BLANKS,
ATTRREL NULLIF ATTRREL = BLANKS,
RETIRED NULLIF RETIRED = BLANKS,
SRN NULLIF SRN = BLANKS,
SOURCE NULLIF SOURCE = BLANKS,
SHAPE_LENG NULLIF SHAPE_LENG = BLANKS,
NAME NULLIF NAME = BLANKS,
TEXTNOTE NULLIF TEXTNOTE = BLANKS,
FEATTYPE NULLIF FEATTYPE = BLANKS,
PLANACC NULLIF PLANACC = BLANKS,
SYMBOL NULLIF SYMBOL = BLANKS,
CREATED NULLIF CREATED = BLANKS,
PEANO,
GEOM COLUMN OBJECT
(
SDO_GTYPE INTEGER EXTERNAL,
SDO_SRID INTEGER EXTERNAL,
SDO_ELEM_INFO VARRAY TERMINATED BY '|/'
(X FLOAT EXTERNAL),
SDO_ORDINATES VARRAY TERMINATED BY '|/'
(X FLOAT EXTERNAL)
)
)
)
LOCATION (TEMP_DATA_DIRECTORY:'GeoData250K_Tas_roads.dat'))
REJECT LIMIT UNLIMITED
Error at Command Line:24 Column:1
Error report:
SQL Error: ORA-30656: column type not supported on external organized table
30656.0000 - "column type not supported on external organized table"
*Cause: Attempt to create an external organized table with a column
of type LONG, LOB, BFILE, ADT, or VARRAY.
*Action: These column types are not supported, change the DDL.
Note, as per the documentation, “These column types are not supported”. For non-point geometry data you must use other methods for loading.
I hope this is useful to someone.


















<<Calling the Oracle Spatial shapefile loader from within the Oracle database itself >>Loading and Processing GPX 1.1 files using Oracle XMLDB
Hello, I am looking for list of companies worldwide who are feeding lidar data to Oracle Spatial.
It would be great if you could help me out. New to Oracle Spacial and need to do a research on how this product is being exploited by companies to provide better services and how successful they are doing so.Thanks,
Victor Nair
— Victor Nair Sep 20, 12:48 am #
Victor,
I would suggest you ask you question on the Oracle Spatial Forum
I have heard of people loading LiDAR into Oracle but I cannot see what business value it would have to a lot of organisations. In most of the work I have been involved with, the processed results of LiDAR data eg a Kriged surface represented as a 2D grid is of more use inside an Oracle database as it can then be combined with vector data, such as a polygon defining a business entity, to calculate new variables or measures such as average elevation etc.
Sorry I can not help you in any other way.
regards
Simon
— Simon Sep 20, 07:50 pm #