Oracle Spatial

A brief guide to Oracle Spatial Extension Simone Campora (simone.campora@epfl.ch)

Spatial Databases A spatial database is a database that is optimized to store and query data related to objects in space, including points, lines and polygons. Up to now the market proposes:

Oracle Spatial Extension (11g r1) Forms a separately-licensed option component of the Oracle Database to aid users in managing geographic and location-data natively Oracle 4 Oracle 7 Oracle 8 Oracle 10 First spatial extension Spatial Data Option (SDO) Oracle Spatial R-Tree indexing Network Model Raster handling Topologies Geocoding

Supported Data Formats While managing spatial data you might have to import data from heterogeneous sources Different GIS (Geographic Information Systems) softwares might support different data types Visit http://www.safe.com/products/desktop/formats/index.php for a complete list data sources – supported types

Oracle Hierarchical Model Oracle datatypes have a hierarchical model

Geometries Geometries refers to spatial objects Their distribution is not uniform but if follows the spatial location of real entities

Spatial Indexing Special indexes are needed to face performant storage and query of geometries Most famous are R-Tree Quad-Tree

R - Tree R-trees are tree data structures that are similar to B-trees, but are used for indexing multi-dimensional information; A common real-world usage for an R-tree might be: "Find all museums within 2 kilometres (1.2 mi) of my current location". The data structure splits space with hierarchically nested, and possibly overlapping, minimum bounding rectangles (MBRs, otherwise known as bounding boxes, i.e. "rectangle", what the "R" in R-tree stands for).

Quadtree A quadtree is a tree data structure in which each internal node has up to four children. Quadtrees are most often used to partition a two dimensional space by recursively subdividing it into four quadrants or regions. The regions may be square or rectangular, or may have arbitrary shapes.

Spatial Query Processing Optional Filter Spatial Data Layer Primary Filter: Spatial Index Data Selection Secondary Filter: Spatial Functions Results

SDO_GEOMETRY Is composed of CREATE TABLE forest( name VARCHAR2(30), animal_population NUMBER(9), geom MDSYS.SDO_GEOMETRY); Methods GET_DIMS() :NUMBER GET_GTYPE :NUMBER

SDO_GTYPE It defines the geometry type stored into the SDO_GEOMETRY object Geometry Type Dimensions Name 2D 3D 4D UNKNOWN_GEOMETRY 2000 3000 4000 POINT 2001 3001 4001 LINESTRING 2002 3002 4002 POLYGON 2003 3003 4003 COLLECTION 2004 3004 4004 MULTIPOINT 2005 3005 4005 MULTILINESTRING 2006 3006 4006 MULTIPOLYGON 2007 3007 4007 NOT OGC

SDO_POINT Used to store Point data It is ignored if SDO_ELEM_INFO and SDO_ORDINATES are not NULL It is of type SDO_POINT_TYPE X NUMBER Y NUMBER Z NUMBER Example: GPS Receiver

SDO_ELEM_INFO It is store as a SDO_ELEM_INFO_ARRAY: VARRAY (1048576) OF NUMBER Its entries must be considered grouping elements by 3 Starting position of the coordinates (starting from 1) Element type (as shown in the table) Interpretation Number Element Type Interpretation 0 UNKNOWN_ELEMENT 1 POINT Number of points of the collection 2 LINESTRING Lines 3 POLYGON 1 lines 1003 External 2 arcs 2003 Internal 3 optimized rectangle 4 circle 4 COMPOUND LINESTRING Number of elements 5 COMPOUNG POLYGON Number of elements 1005 External 2005 Internal

SDO_ORDINATES Stored as SDO_ORDINATE_ARRAY VARRAY (1048576) OF NUMBER It contains a list of object’s coordinates

Examples 1/5 – inserting a Point INSERT INTO pollution VALUES ( 34.6,57.4,etc…, MDSYS.SDO_GEOMETRY( 3001, 352257, MDSYS.SDO_POINT_TYPE(521030,240120,550), null,null) ); SDO_POINT SDO_ELEM_INFO SDO_ORDINATES SDO_GTYPE SDO_SRID

Examples 2/5 – inserting a line INSERT INTO roads VALUES (‘motorway’,’NYC’,etc…, mdsys.sdo_geometry( 2002,352257,null, mdsys.sdo_elem_info_array(1,2,1), mdsys.sdo_ordinate_array(10,10, 20,25, 30,10, 40,10)) ); 10,10 20,25 30,10 40,10

Examples 3/5 – inserting a polygon INSERT INTO roads VALUES (‘motorway’,’NYC’,etc…, mdsys.sdo_geometry( 2003,352257,null, mdsys.sdo_elem_info_array(1,1003,1, 13,2003,3), mdsys.sdo_ordinate_array(2,10, 3,0, 10,0, 12,10, 5,12, 2,10, 4,3, 6,9)) ); (2,10) (2,10) (3,0) (10,0) (12,10) (5,12) (4,3) (6,9)

Oracle Spatial Coordinates System

Projections Projected and Georeferenced 2D projection of the global sphere Cartesian projections Example : Swiss projection Not Georeferenced Local systems not linked to the earth global representation Cartesian projections Ex ample: «non-earth» of MapInfo GIS software Georeferenced not projected Not projected on a spefici plane Latitude Longitude positioning system Example: Oracle, decmal degrees –180 to 180 and – 90 to 90 Example : WGS84

SDO_SRID It is a INTEGER number that identifies which coordinate system is used by the SDO_GEOMETRY The table MDSYS.CS_SRS is defining such reference systems. Description Type Description CS_NAME VARCHAR2(68) Short Name SRID NUMBER(38) Reference System ID AUTH_SRID NUMBER(38) ‘ Oracle’ AUTH_NAME VARCHAR2(256) SRID name WKTEXT VARCHAR2(2046) OGC name CS_BOUNDS MDSYS.SDO_GEOMETRY Can be null

How to Change the SRID? 1. Drop Spatial Indexes drop index town_sidx; 2. Update metadata table update user_sdo_geom_metadata set srid = 352257 where table_name = ‘towns’ and column_name = ‘geom’; 3. Update the SDO_SRID of each object update towns t set t.geom.sdo_srid = 352257; 4. Adjust tolerances update user_sdo_geom_metadata u set u.diminfo = mdsys.sdo_dim_array( mdsys.sdo_dim_element(‘x’,400000,600000,0.005), mdsys.sdo_dim_element(‘y’,100000,300000,0.005)), where table_name = ‘towns’ and column_name = ‘geom’ 5. Recreate Spatial Indexes create index town_idx on towns(geom) indextype is mdsys.spatial_index;

Spatial Operators Make use of spatial indexes They MUST have a spatial index defined at least for the first spatial type used They are only used within the WHERE clause

SDO_FILTER SDO_FILTER(<geometry-1>,<geometry-2>,‘QUERYTYPE=WINDOW’) = ‘TRUE’; geometry-1 (searched object) Must be a column table Must be a SDO_GEOMETRY Must be indexed geometry-2 Must be a variable or column table Must be SDO_GEOMETRY ‘ QUERYTYPE = WINDOW’ Necessary parameter Oracle advises to use only WINDOW select s.nom, s.type from sondages s Where sdo_filter ( s.geom, mdsys.sdo_geometry (2003,352257,null, mdsys.sdo_elem_info_array(1,1003,3), mdsys.sdo_ordinate_array(540300,212500,541300,213500)), ‘ querytype=WINDOW’) = ‘TRUE’;

SDO_RELATE SDO_RELATE(<geometry-1>,<geometry-2>, ‘MASK=<mask> QUERYTYPE=WINDOW’) = ‘TRUE’; Used similarly to SDO_FILTER You can use MASK keywords linked by «+» that means OR operation For example: COVEREDBY, TOUCH, INSIDE, ANYINTERACT select t1.name, t1.na,e from towns t1, towns t2 where c2.name = ‘NYC’ and sdo_relate ( t1.geom,t2.geom, ‘ mask=TOUCH querytype=WINDOW’)=‘TRUE’;

SDO_WITHIN_DISTANCE SDO_WITHIN_DISTANCE( <geometry-1>,<geometry-2>, ‘DISTANCE=<n>,[optional parameters]’) = ‘TRUE’; geometry-2 A buffer will be created starting from this object DISTANCE Distance to be considered QUERYTYPE (optional) Using a primary and secondaryfilter Si QUERYTYPE=FILTER: uses a primary filter only UNIT (optionnel) Defines a specific unit measure that can be different from the SRS definition select a.name from buildings b, streets a where b.id= 2198376 and sdo_within_distance ( a.geom,b.geom,‘distance=1 UNIT=kilometer’)=‘TRUE’;

SDO_NN (nearest neighbor) SDO_NN(<geometry-1>,<geometry-2>,‘[paramètres optionnels]’) = ‘TRUE’; SDO_NUM_RES (optional) number of neighbors to return (default = 1) SDO_BATCH_SIZE (optional) To be used instead of SDO_NUM_RES: it returns subsets of nearest neighbors until all the criteria defined by the « where» condition are satisfied. UNIT (optionnel avec l’utilisation de SDO_NN_DISTANCE) Measure unit select s.type from streets s, polls p where s.id = 1289 and sdo_nn(s.geom,p.geom,’sdo_num_res=5 unit=meter’,1) = ‘TRUE’;

Spatial Functions Are not requiring spatial indexes Can be used on small tables without indexing Can be used either in the « SELECT» or « WHERE » clauses. The input geometries must have the same reference system

SDO_GEOM.RELATE SDO_GEOM.RELATE( <geometry-1>,’<mask>’,<geometry-2>,<tolerance>); geometry-1 (searched object) SDO_GEOMETRY object geometry-2 SDO_GEOMETRY object mask Spatial Relationships to be tested tolerance Tolérance on the data granularity REMINDER: use SDO_RELATE in the where clause because SDO_GEOM.RELATE doesn’t use spatial indexing: it is slower select c.name, sdo_geom.relate(d.geom,’determine’,t.geom,0.5) relation from districts d, towns t where d.distname = ‘NY’ and d.distname = t.district;

Spatial Analysis Oracle spatial functionalities are used to analyze spatial objects The following examples are using spatial functions and operators in order to compute some basic analysis

Calculate a Surface SDO_GEOM.SDO_AREA (<geometry>,<tolerance> [, <unit>]) geometry SDO_GEOMETRY that defines a polygon unit Unit of the results: for a list of possible units look at: select sdo_unit from mdsys.sdo_dist_units; select sum(sdo_geom.sdo_area(c1.geom,0.5,’unit=sq_meter’)) surface from towns t1, towns t2 where t2.name = ‘New York’ and sdo_relate (t1.geom,t2.geom, ’ mask=TOUCH querytype=WINDOW’) = ‘TRUE’;

Calculate the Length SDO_GEOM.SDO_LENGTH (<geometry>,<tolerance> [, <unit>]) Calculate the Distance SDO_GEOM.SDO_DISTANCE (<geometry-1>, <geometry-2>,<tolerance> [, <unit>]) select name,sdo_geom.sdo_length(geom,0.5,’unit=kilometer’) length from streets where number = 1234; select sdo_geom.sdo_distance(f1.geom,f2.geom,0.5) distance from forests f1, forests f2 where f1.id = 1 and f2.id = 2;

Buffer Calculation SDO_GEOM.SDO_BUFFER (<geometry>,<distance>, <tolerance> [, ‘<params>’]) geometry SDO_GEOMETRY distance Distance of the buffer zone Returns: SDO_GEOMETRY select sdo_geom.sdo_buffer (geom,25,0.5,’unit=km’) buffer from streets where number = 120;

Logic Functions SDO_GEOM.SDO_XXX (<geometry-1>,<geometry-2>,<tolerance>) SDO_GEOM.SDO_UNION SDO_GEOM.SDO_INTERSECTION SDO_GEOM.SDO_DIFFERENCE SDO_GEOM.SDO_XOR

References http://en.wikipedia.org/wiki/Oracle_Spatial http://fdo.osgeo.org/ http://download.oracle.com/docs/cd/B28359_01/appdev.111/b28400/sdo_intro.htm Marc Riedo GIS Technology I - EPFL cours slides

Oracle Spatial

More Related Content

Viewers also liked (15)

Similar to Oracle Spatial (20)

Recently uploaded (20)

Oracle Spatial