Location Intelligence

Geographic Information Systems

MapInfo Data Access

Library
Version 1.0

Developer Guide
Information in this document is subject to change without notice and does not represent a commitment on the part of the vendor or its
representatives. No part of this document may be reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopying, without the written permission of Pitney Bowes Inc., One Global View, Troy, New York 12180-8399.
© 2019 Pitney Bowes Inc. All rights reserved.Pitney Bowes Inc. Pitney Bowes, the Corporate logo, MapInfo, Group 1 Software, and MDAL
are trademarks of Pitney Bowes Inc. or a subsidiary. All other trademarks are the property of their respective owners.
Contact information for all Pitney Bowes Inc. offices is located at:: www.pitneybowes.com
This product contains SpatiaLite v 3.1.0, which is licensed under GNU Lesser General Public License, Version 2.1, February 1999. The
license can be downloaded from: http://www.gnu.org/licenses/lgpl-2.1.html. The source code for this software is available from http://
www.gaia-gis.it/gaia-sins/win-binx86-test/spatialite-3.1.0b-test-win-x86.zip and http://www.gaia-gis.it/gaia-sins/win-bin-amd64-test/spatialite-
3.1.0b-test-win-amd64.zip.
This product contains Feature Data Objects v 3.6.0, which is licensed under GNU Lesser General Public License, Version 2.1, February
1999. The license can be downloaded from: http://fdo.osgeo.org/lgpl.html. The source code for this software is available from http://
fdo.osgeo.org/content/fdo-360-downloads.
This product contains HelpLibraryManagerLauncher.exe v 1.0.0.1, which is licensed under Microsoft Public License. The license can be
downloaded from: http://shfb.codeplex.com/license. The source code for this software is available from http://shfb.codeplex.com.
 Table of Contents

1 - Introduction to MapInfo Data Access Selection Class 56

Library 5 SelectionChangedEvent 57
Selection Code Examples 57
Selecting Features Within Another Feature 57
Overview of MapInfo Data Access Library
Checking a Table for Selections 57
(MDAL) 6
Returning All Columns From a Table 58
Table Properties and Methods 6
Event Arguments 58
Samples and Examples 7
Exceptions 59

2 - Working with Data 8 4 - Creating Expressions 60

Overview of MapInfo.Data Namespace 9
Expressions Overview 61
Catalog and Tables 10
Creating Expressions 61
Tables 10
Catalog 15 Where Clause – Boolean Expressions 62
Supported Table Types 16 Functions In Expressions 62
DateTime and Time Expressions 63
Working with Catalog and Tables 20
Locating Open Tables 20 Expression Examples 63
Closing a Table 21
Packing a Table 21 5 - Accessing Data from a DBMS 66
Listening to Table and Catalog Events 23
Table Metadata (TableInfo) 23 Accessing Remote Spatial Data 67
Examining TAB File Metadata 24 Accessing Remote Tables Through a .TAB File
Creating a New Table 25
67
Adding Expression Columns to a Table 28
Accessing Remote Tables Without a .TAB File
Data Sources 29
Choosing the Correct Data Source 30 67
Methods for Accessing Data 30 Mapping DBMS Data with X/Y Columns 68
Data Readers, MemTables and Result Sets 31 Accessing Data from Oracle 68
Using an ADO.NET Data Provider 32 Geometry Conversion 68
Data Binding 35 Oracle Support for Z and M Values 70
Making Tables Mappable 38 SDO_GEOMETRY Arc and Circle Translation70
MapInfo ADO.NET Data Provider 41 Visualization of Non-translatable Oracle Objects
MIConnection 41 71
MICommand 42 Centroid Support 71
MIDataReader 44 Oracle Spatial Reference Support (SRID) 71
MapInfo SQL 45 Accessing Data from MS SQL Server 72
Features and Feature Collections 46 SQL Server 2008 Support 72
Feature 46 DBMS Connection String Format 75
Feature Collections 47 ODBC Connection String Format 75
Searching for Features 47 Oracle Spatial Connection String Format 76
Catalog Search Methods 48 Sample Connection Strings 76
SearchInfo and SearchInfoFactory 49 Defining Mappable Tables in Server Table
Analyzing Data 50 Queries 77
Improving Data Access Performance 52 The Geometry Column 77
The Key Column(s) 78
Accessing Attribute Data 79
3 - Working with Core MDAL Classes54 Performance Issues 79
Working with the Cache 80
Session Interface 55
What Is the Cache? 80
Using Session.Dispose Method 55 How the Cache Works 80

MapInfo Data Access1.0 3 Developer Guide

MapInfo Data Access1.0 Developer Guide
The TableInfoServer Object and the
CacheSettings Property 81
Cache Storage Type: 83
The MapInfo_MapCatalog 83
Loading Spatial Data to DBMS 84
Manually Creating a MapInfo MapCatalog 84
Adding Rows to the MapInfo_MapCatalog 86
Per-Record Styles 91
Symbol, Pen, Brush Clause Syntax 91
Text Objects Limitation 92
Troubleshooting 93

6 - Spatial Objects and Coordinate

Systems 94
Introduction to MapInfo.Geometry Namespace
95
Geometries 95
Geometry Objects 96
FeatureGeometry Objects 97
Geometry Objects 101
Checking for Points in Polygons 103
Coordinate Systems 104
Creating a CoordSys Object 104
Changing the Coordinate System of a Geometry
Object 105
Adding Coordinate Systems 106

7 - Working with GeoPackage 110

Overview 111
Opening a GeoPackage file 111
Opening a GeoPackage Tab file 112
Enable GeoPackage as cache for RDB (SQL/
Oracle) tables 112
Create and Save GeoPackage file
programmatically 113

MapInfo Data Access1.0 4 Developer Guide

1 – Introduction to MapInfo
1
Data Access Library

Welcome Developers to Pitney Bowes’s latest offering for MapInfo Pro

extensibility programming support for Microsoft’s .NET Framework for
Windows. The MapInfo Data Access Library (MDAL) reflects a single
object model for developing Add-in extensions for MapInfo Pro.
The MDAL delivers the Data Access functionality of the MapXtreme
SDK to the MapInfo Pro Add-in developer, providing a robust object
model for creating, accessing and modifying MapInfo spatial tables
from within the .NET Add-in. This provides the organization the
flexibility of developing a spatial tool in .NET that integrates with
MapInfo Pro.

In this chapter:
 Overview of MapInfo Data Access Library (MDAL) . . . . . . . . . . . . . . 6
 Table Properties and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
 Samples and Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
MapInfo Data Access Library 1.0 Developer Guide

Overview of MapInfo Data Access Library (MDAL)

The MapInfo Data Access Library (MDAL) is a set of classes and interfaces that allow
.NET developers to create MapInfo Pro add-ins that can easily create, search, and
update MapInfo Tables and other supported database formats such as Oracle, SQL
Server and GeoPackage. The Library is based on a subset of the MapInfo MapXtreme ™
SDK product, mainly the Data Access functionality

 Currently the MapInfo Data Access Library is only available for use with MapInfo
Pro add-ins.

The following components and features are included in MDAL:

• Fully Capable Data Access Object Model – Create MapInfo Tables, Insert, Update,
Delete, Select, Join multiple tables, Search using a well thought out API.
• Full MiSql Support – Note that there are differences from MapBasic Syntax.
• Complete Geometry Object Model – Supports efficient reading, creation and editing of
all MapInfo Geometry types, including text objects. Also supports conversion to and
from Well Known Binary, Well Known Text and GeoJson formats.
• Full Coordinate System Support.
• Thread-Safe – Can be used to create background tasks in MapInfo Pro or run
processing on secondary threads.
• Supports Most Pro Data formats – not all. Ex: No spatial support for postgis
• Pro style transactions on MapInfo and MapInfo Extended tables - This allows for
background threads to edit tables open in Pro and let the end user decide to commit or
revert the changes.
• Extensive Documentation – API Reference Guide, User Guide, MiSql reference.

Samples and Examples

There are code snippets in the help topics to demonstrate how to use the new types.
Refer to the Sample Applications installed with MapBasic. You can run
MapInfoDataAccessLibraryExamples to create a window that lets you click on an
example in the User Interface for a property or methods. It then takes you to the debugger
in the sample code.
Samples are placed at:
SAMPLES\RIBBONINTERFACE\DotNet\MapInfoDataAccessLibraryExamples.

Pitney Bowes 6 Developer Guide

 1 – Introduction to MapInfo Data Access Library

Technical Support
Pitney Bowes offers unparalleled technical support for users of MapInfo software
products. Our Technical Support department provides technical assistance to registered
users of MapInfo software – so you don't need to be an expert in all aspects of our
products in order to get results. See the Pitney Bowes Web site at www.pb.com/software
for information on the tech support offerings.

Pitney Bowes 7 Developer Guide

2 – Working with Data
2
The MapInfo.Data namespace contains the classes and interfaces that
provide multiple ways of accessing data from a MapInfo Data Access
Library (MDAL) application.

In this chapter:
 Overview of MapInfo.Data Namespace . . . . . . . . . . . . . . . . . . . . . . . 9
 Catalog and Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
 Supported Table Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
 Working with Catalog and Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
 Table Metadata (TableInfo) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
 MapInfo ADO.NET Data Provider . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
 Features and Feature Collections. . . . . . . . . . . . . . . . . . . . . . . . . . . 46
 Analyzing Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
 Improving Data Access Performance . . . . . . . . . . . . . . . . . . . . . . . . 52
 2 – Working with Data

Overview of MapInfo.Data Namespace

The MapInfo.Data namespace contains the classes and interfaces that provide multiple
ways of accessing data from a MapInfo Data Access Library (MDAL) application. Within
this namespace is the MapInfo ADO.NET data provider with a MapInfo SQL language for
standard querying of databases and tables. The Feature object model is another way to
access data that uses objects instead of SQL. The Catalog is the starting point for data
access, containing methods for managing tables (open, close, create) and searching for
data in a variety of ways.

This chapter is organized to follow the MapInfo Data Access Library (MDAL) Data Access
Model diagram above, and includes these topics:
• Catalog and Tables
• Supported Table Types
• Table Metadata (TableInfo)
• MapInfo ADO.NET Data Provider
• Features and Feature Collections

Pitney Bowes 9 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Data access is central to any MDAL application, and covers a wide variety of topics.
Within the topics listed above are other important topics of information that should not be
overlooked.
Following this chapter are two additional chapters related to data access: Chapter 4
Creating Expressions, and Chapter 5 Accessing Data from a DBMS.

Catalog and Tables

Catalog is the manager of the MapInfo Data Access Library (MDAL) data access model.
Tables are a fundamental unit of MapInfo Data Access Library (MDAL). Tables hold the
data that you want to display and analyze in your application. The Catalog, as manager,
holds a list of tables that are currently open in the session. Tables are also opened,
created and closed from the Catalog.
Nearly all of MDAL’s data access operations involve the Catalog and tables.

Tables
The Table class is the basic unit of all data access. Table, Column, and all TAB file
metadata information is accessible from a MapInfo Table. Tables may be mappable
(contain a column of type FeatureGeometry) or non-mappable.

Pitney Bowes 10 Developer Guide

 2 – Working with Data

Table Aliases
When tables are opened, they can be assigned a name (or alias) which is used while the
table is open for referencing the table. For example, the table may be referred to by its
alias in SQL statements. A table that is opened from a TAB file is assigned a default alias
if no alias is specified. The default alias is based upon the name of the TAB file. This
property is optional and may be set to null. However, it is good practice to assign an alias.

Columns
A Column object identifies the properties of a column in a table, feature, or feature
collection and specifies the column's name (alias), data type, width (for string and decimal
columns), and other properties of the column.
Supported data types include:

Data Type Description

Int Provides a 32-bit signed integer. This maps to the .NET

Framework datatype Int32.

SmallInt Provides a 16-bit signed integer. This maps to the .NET

Framework datatype Int16.

Double A floating point number within the range of -1.79E +308 through
1.79E +308. This maps to Double.

dBaseDecimal Provides a floating point number which is treated internally the

same as a Double. The dBaseDecimal has a fixed precision
and scale when persisted in a table. This is a legacy data type
derived, as its name suggests, from the dBase file formats. This
maps to Double.

Boolean Provides a boolean value. This maps to Boolean.

String Provides a variable-length, null terminated UNICODE string

value. This maps to String.

Date * † Provides a date value. The Date type is implemented as a

structure in the MapInfo.Data namespace.

DateTime * ‡ Provides a combined date and time value. The DateTime type
is mapped to System.DateTime.

Pitney Bowes 11 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Data Type Description

Time Provides a time value. Supports the Time type in MapInfo

Professional tables (TAB files) version 9.0 and later. The Time
type is implemented as a structure in the MapInfo.Data
namespace.

FeatureGeom Provides a FeatureGeometry.

etry

Binary Provides an array of binary data. This maps to an Array of Byte

values.

Key Provides a key from a table. This is the data type of the Key
pseudo column on a Table.

CoordSys Provides a coordinate system. This type exists only for the
purposes of binding a coordinate system object to an
MICommand for functions which require the specification of a
coordinate system.

Style Provides an instance of a Style class. See MapInfo.Styles.Style.

This is the data type of the Style object stored in the style
column on a Table.

Raster Provides a RasterInfo from table's raster column. This is the

data type of the RasterInfo object stored in raster column on a
Table.

Grid A GridInfo from table's grid column. This is the data type of the
GridInfo object stored in grid column on a Table.

Wms Provides a WmsClient from the table's Wms column. This is the
data type of the WmsClient object stored in the Wms column on
a Table.

TileServer Provides a TileServerInfo from the table’s raster column. This is

the data type of the TileServerInfo object stored in the raster
column on a Table.
* To ensure backward compatibility with earlier versions of MDAL, the
MapInfo.Data.MIDataReader.GetDateTime method works with both the DateTime and Date types. In
both cases, a System.DateTime value is returned. However, the MapInfo.Data.Column.DataType will
reflect the actual data type, either Date or DateTime.

Pitney Bowes 12 Developer Guide

 2 – Working with Data

† The Time and DateTime types are not supported for MapInfo SQL functions. However, in MapInfo
SQL functions that call/use a DateTime type, the function will return the date portion of the DateTime
value. Please see the MapInfo SQL Reference for more information.
‡ The Time and DateTime types are not supported for MapInfo SQL functions. However, in MapInfo
SQL functions that call/use a DateTime type, the function will return the date portion of the DateTime
value. Please see the MapInfo SQL Reference for more information.

Time and DateTime Data Source Support

MapInfo Data Access Library (MDAL) can read Date, DateTime, and Time data (and save
it back if applied) on the supported data sources and data providers. The different data
sources may have different type definitions on date/time, which may or may not match
MapInfo Data Access Library (MDAL) types exactly.
The new data types are supported for the following data sources:
• Mem tables
• Native tables (TAB files)
• ADO.NET
• Oracle via OCI
• MS SQL Server via ODBC
The ASCII and dBase, and Microsoft Access data sources are not supported.

Date and DateTime Support in Remote Databases

Remote databases may not support all the data types that MapInfo Data Access Library
(MDAL) supports. The table below shows the date and time-based types supported in
native TAB files and in each supported database.

Native(X) MS SQL
MDAL TAB Files ADO.NET Oracle (OCI) Server GeoPackage

Date Date Date

Time Time

DateTime DateTime DateTime DateTime DateTime DateTime

The following sections provide you with MI_Key, MI_Geometry, and MI_Style column
information.

Pitney Bowes 13 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

MI_Key

All tables have a pseudo column named MI_Key which returns instances of Key. The
MI_Key pseudo column is similar in concept to the rowid pseudo column in MapInfo
Professional and MapBasic. Unlike rowid, this column is not a numeric column. A Key
instance may be converted to or from a string literal.

MI_Geometry

A Geometry column object in a table, feature, or feature collection contains

FeatureGeometry objects and specifies properties such as the coordinate system of the
column and the entire bounds of all the geometry objects it contains.
Geometry columns for most table types are given the name “Obj”. To be compatible with
previous versions of MapX and MapInfo Professional, the alias “Obj” is resolved to the
first GeometryColumn in the table. Additionally, the alias “MI_Geometry” may also be
used for any table to refer to the same column that “Obj” refers to.

MI_Style

Tables with a Geometry column also have a column with the name “MI_Style”, or if not
found, from the first column with type MIDbType.Style. This column is used to hold the
style information for Geometry objects such as line width for polygons and symbol size for
points. This column cannot be updated independently. The Style and Geometry columns
must be updated at the same time.
The MI_Style column is created automatically when you are opening a table in MapInfo
native format (.TAB). For all other table types, you must specifically create the column. If
you use MapInfo.Data.ColumnFactory.CreateStyleColumn it will create a column with the
name (alias) of "MI_Style" and a data type of MIDbType.Style.
When using MISQL to insert rows into a table, be sure to include the MI_Style column in
the insert statement. See the code example below:
Table tab = MapInfo.Engine.Session.Current.Catalog.GetTable("MapViewer");
TableInfo ti = TableInfoFactory.CreateTemp("Test",
((MapInfo.Data.GeometryColumn)tab.TableInfo.Columns["Obj"]).CoordSys);
Table tabTemp = MapInfo.Engine.Session.Current.Catalog.CreateTable(ti);

MIConnection conn = new MIConnection();

conn.Open();
MICommand comm = conn.CreateCommand();
comm.CommandText = "Insert Into " + tabTemp.Alias +
" (Obj, MI_Style) SELECT MI_Point(MI_X(Obj), MI_Y(Obj), '" +
((MapInfo.Data.GeometryColumn)tab.TableInfo.Columns["Obj"]).Coord
Sys.SrsString + "'), MI_Style" + " FROM " + tab.Alias + " WHERE msaname
= 'Minneapolis-St. Paul, MN-WI' AND Not Obj = Null";
MessageBox.Show(comm.CommandText);

Pitney Bowes 14 Developer Guide

 2 – Working with Data

int numChanged = comm.ExecuteNonQuery();

mapControl1.Map.Layers.Add(new FeatureLayer(tabTemp));
mapControl1.Map.SetView(mapControl1.Map.Layers["Test"] as FeatureLayer);

Catalog
The Catalog is essentially the manager of the MapInfo Data Access Library (MDAL) data
access model. The Catalog holds a list of tables that are currently open in the MDAL
Session. Tables are also opened, created and closed from the Catalog. The Catalog can
be thought of as a single database holding all the tables opened in it, regardless of their
actual data source.
Each MDAL Session manages a single Catalog.
Catalog initially contains no tables. When a table is opened, an alias (or name) is
assigned to the table or provided by the caller. The alias is used to identify the table in
queries and other operations.
Tables can be mappable (contain a spatial component) or be non-mappable and contain
only data columns. The MapInfo Data Access Library (MDAL) Catalog can open both
types and use either in queries and joins.
Catalog provides facilities for creating new table definitions and enumerating through
tables which are currently opened. Catalog also contains search methods that can be
used to access data in open tables.
The Catalog has an SQL engine that allows you to select, insert, update, and delete
tables and data within tables. The SQL engine allows you to join any tables defined in the
catalog (for example, Native to SQLServer, or SQLServer to Oracle). The Catalog
handles the integration from various sources so you don’t have to. This is a powerful tool
when organizing data from various sources.
The MDAL Catalog is exposed through the MapInfo ADO.NET Data Provider. Access to
tables and result sets is controlled through this interface. See MapInfo ADO.NET Data
Provider.

Code Sample
The following example illustrates how to access the Catalog through the MDAL Session
object, open some tables and enumerate through all the tables in the Catalog followed by
only the editable tables in the Catalog.
VB example:
Public Shared Sub MapInfo_Data_Catalog()
’ Catalog is accessible off the Session object

Pitney Bowes 15 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Dim catalog As Catalog = Session.Current.Catalog

’ Open a bunch of tables

Dim table As Table = catalog.OpenTable("States.tab")
table.SessionInfo.ReadOnly = True ’ Make states ReadOnly
table = catalog.OpenTable("world.tab")
table = catalog.OpenTable("worldcap.tab", "World Capitals")

’ Enumerate the catalog directly - includes All tables

Dim t As Table
For Each t In catalog
Console.Out.WriteLine("Table : {0}", t.Alias)
Next
Console.Out.WriteLine()

’ Now enumerate through only tables that are editable (not ReadOnly)
Dim tEnum As ITableEnumerator = _
catalog.EnumerateTables(TableFilterFactory.FilterEditableTables())
While tEnum.MoveNext()
Console.Out.WriteLine("Table: {0}", tEnum.Current.Alias)
End While

Session.Current.Catalog.CloseAll()
End Sub

Supported Table Types

One of the strengths of MapInfo Data Access Library (MDAL) is its ability to access data
"where it lives." This means we strive to handle a wide variety of data formats. Here are
the supported table types in MapInfo Data Access Library (MDAL):

MapInfo .TAB MapInfo native table format.

format
This file-based table may have an associated .MAP file
containing FeatureGeometry and Style information. Non-
spatial data is stored in a .DAT file. TAB is available as a
storage format to be used when caching. See Creating a
New Table.

dBase Data stored in a dBase file.

The table may have an associated .MAP file containing
FeatureGeometry and Style information. Non-spatial data is
stored in a .DBF file. An associated .IND file holds one or
more B-Tree indices for non-spatial attribute values (strings,
numbers, and dates)

Pitney Bowes 16 Developer Guide

 2 – Working with Data

ASCII Data stored in a delimited .CSV or text file. The maximum

string length is 255 characters (including up to two quotation
marks). ASCII tables are Insert only.
The table may have an associated .MAP file containing
FeatureGeometry and Style information. Non-spatial data is
stored in a .CSV or .TXT file.

MS Access Microsoft Access database table.

This file-based table located inside of a Microsoft Access
.MDB database may have an associated .MAP file containing
FeatureGeometry and Style information. Non-spatial data is
stored in an Access file.

Shapefile An ESRI Shapefile table.

These tables are read-only and support three-dimensional
geometries (X, Y, Z, M). Non-spatial attribute data is stored in
.DBF file format. FeatureGeometry values stored in ESRI
.shp file format. MDAL does not have access to the spatial
index. Caching is supported as a .MAP file that can be
temporary or persistent. A persistent cache can be shared
with MapInfo Professional. It is controlled by the
PersistentCache property on the TableInfoShapefile class.

MemTable An in-memory storage of non-spatial attribute data.

FeatureGeometry data and indices are stored on disk. These
are temporary tables—all data is lost when the table is
closed. MemTables are serializable. The data can be
persisted in workspaces (data only; keys may be changed
when reloading). This table type is available as a storage
format to be used when caching. There is no .TAB file
equivalent to a MemTable. See Create a Temporary
MemTable.

Pitney Bowes 17 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

RDBMS Server A spatial table stored in a remote database management

system (such as SQL Server or Oracle).
The table is defined by a native SQL SELECT statement.
MDAL performs query parsing and modification. Caching is
enabled by default. Supported protocols (toolkits) include:
OCI (Oracle Spatial) and ODBC (, SQL Server, SpatialWare
and XY). See Chapter 5 Accessing Data from a DBMS.

ADO.NET A table of non-spatial data that is based upon an ADO.NET

DataTable or IDbCommand.
This table type supports many different data providers with
provider-specific implementations. ADO.NET is the choice
when there is no MDAL supported data provider. ADO.NET is
designed to support both Connected (IDBCommand) and
Disconnected (DataTable) ADO.NET models. IDBTables are
read-only. Cache may be applied forcibly (implicit keys).
DataTables are editable and run-time serialization is
supported. See Using an ADO.NET Data Provider.

Raster A table containing a raster image.

This typically provides a base map for other spatial table
types. Tables have only a single record and a fixed column
schema (RasterInfo, MI_Geometry, MI_Style). These tables
may be joined with vector tables using spatial predicates (for
example, “within”).

Grid A table containing a grid image.

This table type provides a base map for other spatial table
types. Tables have only a single record and a fixed column
schema (GridInfo, MI_Geometry, MI_Style). These tables
may be joined with vector tables using spatial predicates (for
example, “within”). GridRead class provides access to grid
cell values. MapInfo.Raster.GridCreatorFromFeatures class
creates a grid using an interpolator.

Pitney Bowes 18 Developer Guide

 2 – Working with Data

WMS A table containing an image from a Web Map Service (WMS).

This table type provides a base map for other spatial table
types. Tables have only a single record and a fixed column
schema (GridInfo, MI_Geometry, MI_Style). These tables
may be joined with vector tables using spatial predicates (for
example, “within”). WMS tables are accessed like dynamic
raster through a MapInfo.Wms.WmsClient.

Seamless A table that combines two or more base tables with

contiguous geography. It displays as a single map layer.
Seamless tables are specifically tuned for spatial queries,
such as drawing a map, which uses seamless tables for
optimally querying appropriate component tables.
Component tables that make up a seamless table may be
vector or raster. They must all have the same schema. They
are read-only. The underlying component tables cannot be
modified directly. Sorting and aggregating operations
examine every record of every component tables (could be
have a significant performance impact when working with
vector tables.)

View A view based on a MapInfo SQL Select statement (not a

native SQL supported by Server tables). See View Tables.

ResultSet A table containing the results of a search. ResultSet is used

exclusively for IResultSetFeatureCollections. See Result
Sets.

TileServer A table containing a TileServer image. This typically provides

a base map for other spatial table types. Tables contain a
single record and a fixed column schema (TileServerInfo,
MI_Geometry, MI_Style).

Pitney Bowes 19 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Geopackage A table containing information in a Geopackage format.

Indicates a table that has both FeatureGeometry objects and
attribute data stored in an OGC Geopackage database file
format.

NativeX MapInfo Extended (NativeX) Tab file formats.

This file-based table may have an associated .MAP file
containing FeatureGeometry and Style information. Non-
spatial data is stored in a .DAT file. TAB is available as a
storage format to be used when caching.
The NativeX format supports table caches larger than 2GB in
size and character sets UTF-8 and UTF-16.

Working with Catalog and Tables

This section covers some basic table operations, including:
• Locating Open Tables
• Closing a Table
• Packing a Table
• Listening to Table and Catalog Events
See also the MapInfo.Data.Table class in the MapInfo Data Access Library (MDAL)
Developer Reference.

Locating Open Tables

To locate open tables, you must enumerate the catalog. This is done by using the
methods in the following sections.

Catalog.GetTable
The MapInfo.Data.Catalog.GetTable method returns the Table object referenced by the
TableAlias parameter. This must be a table which has already been opened. If no such
table is found (or the table has subsequently been closed), then the method returns null.

Pitney Bowes 20 Developer Guide

 2 – Working with Data

Catalog.Item (Indexer)
MapInfo.Data.Catalog.Item property can be used as an indexer for locating a Table by its
Alias. This is functionally equivalent to using the Catalog.GetTable method but generates
code that is easier to read. The Alias must specify a table which has already been
opened.
VB example:

Public Shared Sub MapInfo_Data_Catalog2()

Dim tbl As Table

For Each tbl In Session.Current.Catalog
System.Console.WriteLine("Table: " + tbl.Alias)
Next

End Sub

TableEnumerators
Table enumerators may be obtained through the various overloaded EnumerateTables
methods. A table enumerator may be created with a filter. The filter determines which
tables are actually included in the enumeration while the enumerator simply provides the
mechanics of enumeration. You can create your own table filters to use in the
TableEnumerator. You can also create your own table enumerator by implementing the
ITableEnumerator interface.
VB example:
Public Shared Sub MapInfo_Data_Catalog3(ByVal catalog As Catalog)
Dim te As ITableEnumerator = _
catalog.EnumerateTables(TableFilterFactory.FilterEditableTables())

While te.MoveNext()
Dim tbl As Table = te.Current

End While
End Sub

Closing a Table
Three methods are available to close tables. MapInfo.Data.Catalog.CloseAll closes all
open tables while Catalog.CloseTable closes a single, open table. The Table class also
has a Close method.

Packing a Table

Pitney Bowes 21 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

The MapInfo.Data.Table.Pack method removes records from the table that were
previously marked for deletion. When the table is packed, the table's TablePacked event
is raised. The arguments for the event indicate whether or not the table's keys changed
as a result of the pack (which would be caused by removing deleted records). Keys only
change if the PackType includes RemoveDeletedRecords and if there actually were
deleted records in the middle of the table. If the only deleted records in the table are at the
end of the table, then no keys are changed. The event does not indicate that keys were
changed.

 Since ResultSet tables hold collections of keys, these are vulnerable to pack
operations on the table from which they were derived. The ResultSet is no longer
valid if the keys have changed.

PackType Enumeration provides you with the following options.

• PackGeometry – Indicates that the geometry objects are packed. Packing the objects
attempts to remove as much unused space as possible. A fully packed RTree (the
spatial index used to spatially access the geometry objects) may reduce performance
by causing many more unnecessary reads. To balance disk space and processing
speed, packing the geometry objects may continue to leave some unused space in the
RTree. Also note: a packed RTree results in a slight performance penalty for insert and
update operations as there is a higher likelihood that the RTree needs to be expanded.
• RebuildGeometry – Rebuilding the geometry objects removes unused space that has
resulted from a series of insert, update, and/or delete operations. Unlike packing the
geometry objects, this option intentionally leaves unused space in the RTree index to
improve the performance of future insert and update operations.
• PackIndex – Non-spatial indices are maintained as B*trees. These structures do not
always have filled internal or leaf nodes. This is intentional by default to allow room for
the index to accommodate insert and update operations without requiring a significant
restructuring of the index. The unused space may be exacerbated by the occurrence
of insert, update, or delete operations. Packing an index fully packs every internal and
leaf node (except possibly the “last” node). This option reduces the disk space used by
the index as much as possible and also improves the read-performance of the index.
There is a performance penalty for insert and update operations on a fully packed
index.
• RebuildIndex – Rebuilding an index does not fully pack the internal and leaf nodes like
the PackIndex option. Instead, rebuilding an index recreates the index with the amount
of unused space that is intentionally put into the index to balance disk space, read
performance, and modify performance. After several modification operations, an index
may contain a considerable amount of unused space. This option regains that unused
space.

Pitney Bowes 22 Developer Guide

 2 – Working with Data

• RemoveDeletedRecords – Some data sources, including MapInfo Native and dBase

data sources, do not physically remove records when they are deleted. To physically
remove the deleted records, the table must be packed with this option specified. The
record number is typically used as the record key for these data source types.
Removing deleted records from a table may cause keys to become invalid since they
may change as a result of the pack.
• CompactDb – If the table's data source is Microsoft Access (TableType of Access),
then the MDB file containing the table's data may also be compressed using the Pack
method and specifying this option.
• All – This is a convenience option that is equivalent to PackGeometry | PackIndex |
RemoveDeletedRecords.

Listening to Table and Catalog Events

Table exposes several events which applications may subscribe to. They are:
• RowInsertedEvent – Occurs when a new row is added to the table.
• RowUpdatedEvent – Occurs when an existing row in the table is updated.
• RowDeletedEvent – Occurs when a row in the table is deleted.
• TablePackedEvent – Occurs when the table is packed.
• TableCloseRequestEvent – Occurs when the table has been asked to close.
• TableIsClosingEvent – Occurs when the table is closing.
• TableClosedEvent – Occurs when the table is closed.
Catalog also exposes the following events.
• TableOpenedEvent – Occurs when a table is opened.
• TableCreatedEvent – Occurs when a new table is created.
• TableIsClosingEvent – Occurs when the table is closing.

Table Metadata (TableInfo)

The TableInfo class in the MapInfo.Data namespace is an abstract base class that
contains information, or metadata, about an existing table, including:
• Columns – number, names, data types, etc.
• Table alias, and description and pathname of the data source.
• Client metadata (the information between the begin_metadata/end_metadata tags in
the TAB file).

Pitney Bowes 23 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

TableInfo is used to open tables and create new tables. It is also used for retrieving
information about the open table.
Classes that derive from TableInfo include provider-specific metadata. There is a
TableInfo implementation for every MapInfo Data Access Library (MDAL) supported table
type. See Data Sources.
TableInfo instances may be constructed manually, or from a .TAB file definition (without
opening the table), as shown below.
TableInfo.CreateFromFile(…)

TableInfo contains properties for enabling Table Services, including caching and making a
table mappable via a spatial schema. See Working with the Cache and Making Tables
Mappable.

MapInfo Data Access Library (MDAL) provides table column metadata support for M and
Z values. This feature is useful when you want to know whether geometries of a particular
data provider can support 3D and Measured values without evaluating its individual
geometries.
Metadata for a table can be accessed from the table's TableInfo property. From the table
metadata you can access the GeometryColumn to interrogate if the table supports M or Z
values and what the range of values for that table is if the range is known. For more
information on support for M and Z values, see Support for M and Z Values.

Examining TAB File Metadata

Pitney Bowes 24 Developer Guide

 2 – Working with Data

TAB file metadata is accessible and editable. The TableInfo class can be obtained from
the Table to get information about the table structure.
The following code demonstrates how to get the metadata for an open table. The code
also demonstrates how the geometry column can be used to determine the coordinate
system and bounds of the table. For a code example that returns M and Z values, see
MapInfo.Data.TableInfo in the Developer Reference.
VB example:
Public Shared Sub MapInfo_Data_TableInfo2()
' Get the metadata for an open table
Dim ti As TableInfo = Session.Current.Catalog("states").TableInfo

' Print out some information to the console

Console.Out.WriteLine("Table Alias={0}, Datasource={1}, _
Description={2}, Type={3}", _
ti.Alias, ti.DataSourceName, ti.Description, ti.TableType)

' Print out some information about each column

Dim col As Column
For Each col In ti.Columns
Console.Out.WriteLine("Column {0} Type={1} Width={2}", _
col.Alias, col.DataType, col.Width)
' If the column is a geometry column, print csys and bounds.
If col.DataType = MIDbType.FeatureGeometry Then
Dim geocol As GeometryColumn = col
Dim csys As MapInfo.GeomeTry.CoordSys = geocol.CoordSys
Console.Out.WriteLine("CSys : {0}", csys.MapBasicString)
Dim dr As MapInfo.GeomeTry.DRect = geocol.Bounds
Console.Out.WriteLine("Bounds=({0},{1}),({2},{3})", dr.x1, _
dr.y1, dr.x2, dr.y2)
End If
Next

End Sub

Creating a New Table

The following sections illustrate how to create a permanent native table, a temporary
native table, and a temporary MemTable.

Create a New Permanent Native Table

The MapInfo.Data.Table.TableInfo property for a MapInfo native table returns an instance
of TableInfoNative. A native table is a MapInfo .TAB file. This class may be used to
access properties that are specific to native table types. New instances of this class may
be created and used to construct new tables. See also Data Sources.

Pitney Bowes 25 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Note the use of the ColumnFactory class. This is provided to help you know which
arguments are necessary for different data types. For example, a geometry column
requires a coordinate system.
VB example:
Public Shared Sub MapInfo_Data_TableInfoNative()
Dim ti As TableInfoNative = New TableInfoNative("NewTable")
ti.TablePath = "c:\data\Capitals.TAB"
ti.Columns.Add(ColumnFactory.CreateIndexedStringColumn("Capital",_
25))
ti.Columns.Add(ColumnFactory.CreateStringColumn("Country", 30))
ti.Columns.Add(ColumnFactory.CreateDoubleColumn("Pop_Grw_Rt"))

' Make the table mappable

ti.Columns.Add(ColumnFactory.CreateStyleColumn())
Dim Robinson As CoordSys = _
Session.Current.CoordSysFactory.CreateFromPrjString("12, _
62, 7, 0")

ti.Columns.Add(ColumnFactory.CreateFeatureGeometryColumn(Robinson))
' Note we do not need to (nor should we) add a column of type Key.
' Every table automatically contains a column named "MI_Key".
Dim table As Table = Session.Current.Catalog.CreateTable(ti)

End Sub

Create a Temporary Native Table

VB example:
Public Shared Sub MapInfo_Data_TableInfo3(ByVal conn As MIConnection)
Dim ti As TableInfoNative = New TableInfoNative("NewTable")
ti.Temporary = True
Dim col As Column

col = New Column

col.Alias = "FString30"
col.DataType = MIDbType.String
col.Indexed = True
col.Width = 30
ti.Columns.Add(col)

col = New Column

col.Alias = "FInt32"
col.DataType = MIDbType.Int
col.Indexed = True
ti.Columns.Add(col)

col = New Column

col.Alias = "FInt16"
col.DataType = MIDbType.SmallInt

Pitney Bowes 26 Developer Guide

 2 – Working with Data

col.Indexed = True
ti.Columns.Add(col)

col = New Column

col.Alias = "FDouble"
col.DataType = MIDbType.Double
ti.Columns.Add(col)

col = New Column

col.Alias = "FDateTime"
col.DataType = MIDbType.Date
ti.Columns.Add(col)

col = New Column

col.Alias = "FBoolean"
col.DataType = MIDbType.Boolean
ti.Columns.Add(col)
’ Note we do not need to (nor should we) add a column of type Key.
’ Every table automatically contains a column named "MI_Key".
Dim miTable As Table = conn.Catalog.CreateTable(ti)
End Sub

Create a Temporary MemTable

The MapInfo.Data.Table.TableInfo property for a memory table returns an instance of
TableInfoMemTable. This class may be used to access properties that are specific to
memory table types. New instances of this class may be created and used to construct
new tables.
Data in formats such as XML or GML from a Web service can be brought into the Catalog
and used in this fashion. It can be converted to a MultiPolygon, LineString, Point, or other
Geometry via the MapInfo Data Access Library (MDAL) API. MDAL then turns the
Geometry into a FeatureCollection, and, in turn, saves it to a memTable or native TAB
format.
This approach is also appropriate if you wish to make data available for use in MDAL, but
not necessarily for map display.
MapInfo Data Access Library (MDAL) supports reading and writing Z and M values to
MemTables. M values on MultiCurves allow you to carry out linear referencing operations
and dynamic segmentation.
VB example:
Public Shared Sub MapInfo_Data_TableInfoMemTable()
Dim ti As TableInfoMemTable = New TableInfoMemTable("NewTable")
' Note: The TablePath property does not apply - it can be set but it _
' is meaningless.

ti.Columns.Add(ColumnFactory.CreateIndexedStringColumn("Capital",_

Pitney Bowes 27 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

25))
ti.Columns.Add(ColumnFactory.CreateStringColumn("Country", 30))
ti.Columns.Add(ColumnFactory.CreateDoubleColumn("Pop_Grw_Rt"))

' Make the table mappable

ti.Columns.Add(ColumnFactory.CreateStyleColumn())
Dim Robinson As CoordSys = _
Session.Current.CoordSysFactory.CreateFromPrjString("12, 62, _
7, 0")
ti.Columns.Add(ColumnFactory.CreateFeatureGeometryColumn(Robinson))
' Note we do not need to (nor should we) add a column of type Key.
' Every table automatically contains a column named "MI_Key".
Dim table As Table = Session.Current.Catalog.CreateTable(ti)
End Sub

Adding Expression Columns to a Table

Use the MapInfo.Data.Table.AddColumns method to add expression columns to a table.
The form of AddColumns that takes a Columns object creates temporary columns based
on expressions comprised of functions, operators, literal values, and other columns on
the table. All instances of Column in the columns argument must have an expression
string specified.

 TableAddColumns is not supported for the following table types: Server, View,
Seamless, AdoNet, ResultSet, or Drilldown. MapInfo Data Access Library (MDAL)
checks for the table and throws an exception if it encounters one of these table
types.

VB example:
Public Shared Sub MapInfo_Data_TableAddColumns(ByVal miTable As Table)
Dim NewCols As Columns = New Columns
NewCols.Add(New Column("PopDensity1990", "Pop_1990 / _
MI_Area(Obj, 'sq mi', 'Spherical')"))
NewCols.Add(New Column("PopDensity2000", "Pop_2000 / _
MI_Area(Obj, 'sq mi', 'Spherical')"))
miTable.AddColumns(NewCols)
End Sub

The expression string “Pop_1990 / MI_Area(Obj, 'sq mi', 'Spherical')” represents derived
information that will be placed in the temporary column. It says ‘For each record divide
population by area in square miles to yield the population density.’ The SQL function
MI_Area () in the expression will derive the area from the geometry of the record.
Using the AddColumns method may offer performance improvements in desktop
applications where the join can be performed once, rather than on each subsequent
access (as in the case of a view).

Pitney Bowes 28 Developer Guide

 2 – Working with Data

For more information and code examples, see the MapInfo.Data.Table.AddColumns class
in the Developer Reference Help system.
For more information on creating expressions, see Chapter 4 Creating Expressions.

Data Sources
The following table lists the data sources supported by MapInfo Data Access Library
(MDAL). Each type of data source is accessed by a specific data provider called TableInfo
class, that is derived from MapInfo.Data.TableInfo. For a short summary of each data type
see Supported Table Types.

Data Source Class

Native (MapInfo.TAB) TableInfoNative

dBase TableInfodBase

MS Access TableInfoMSAccess

ASCII TableInfoAscii

RSBMS Server TableInfoServer

ESRI Shapefile TableInfoShapefile

Seamless TableInfoSeamless

Raster TableInfoRaster

Grid TableInfoGrid

WMS TableInfoWMS

ADONET TableInfoAdoNet

MemTable TableInfoMemTable

View TableInfoView

ResultSet TableInfoResultSet

TileServer TableInfoTileServer

GeoPackage TableInfoGeoPackage

NativeX TableInfoNativeX

Pitney Bowes 29 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Choosing the Correct Data Source

Choosing the correct data source can make a difference in your application’s
performance. In some cases you will not have a choice, such as native MapInfo files
(.TAB), but in other cases there may be multiple choices. In most cases, you will be using
a supported data provider for the data source. In cases where the data is not accessible
through one of these, you may be able to use the MapInfo ADO.NET data provider. This
is the same data provider interface that the Catalog uses to retrieve data.
Each data source has certain performance characteristics. Native tables offer the best
access and map drawing times. Data is stored locally on the system and optimized for
your current operation. Other file-based table types perform well, depending on current
hardware and file size.

Methods for Accessing Data

MapInfo Data Access Library (MDAL) provides several ways to bring data into the
Catalog:
• Direct access to data sources
• Access via an ADO.NET data provider (TableInfoAdoNet)
• XML/GML from third-party web services
The best method to access data is to open it directly using one of the TableInfo classes
that are specific to where your data resides.
Use the second method (TableInfoAdoNet) to access data that is not internally supported
but has an ADO.NET provider.
A third method allows developers to integrate data to the Catalog who may interact with
HTTP services that return XML or GML.

Direct Access to Data Sources

MapInfo Data Access Library (MDAL) provides native support for accessing data stored in
file-based table formats and RDBMS servers, such as SQL Server and Oracle. In the
case of file-based access, provide the path and filename in the appropriate TableInfo
instance (TableInfoNative, TableInfodBase, TableInfoMSAssess, TableInfoAscii,
TableInfoGeopackage, and TableInfoNativeX).
For direct access to data stored in RDBMS serves, use the TableInfoServer class to
define the connection string and an SQL statement to execute on the remote table.
Internally, MDAL uses ODBC or OCI to access the remote database.

Pitney Bowes 30 Developer Guide

 2 – Working with Data

TableInfoServer will open a connection to the server, query the table's metadata, and
create the appropriate table definition with any spatial characteristics that are defined on
the remote server. This tends to be the best performing method with remote data.
Internally, MDAL can access only the data necessary to perform the current operation.
During a map draw, MapInfo Data Access Library (MDAL) will construct a query that
returns only the geometry column, and not the data columns. This minimizes the network
traffic. If caching is on, then this is only an issue for the first access, since all subsequent
requests will come from the cache. See Chapter 5 Accessing Data from a DBMS.

Access via an ADO.NET DataProvider

The second data access method is to use an ADO.NET data provider. This requires the
definition of ADO.NET classes for data retrieval. Only non-mappable tables may be
supplied as an AdoNet table. Non-mappable tables are those that do not contain
geometry information about the data. Tables retrieved from an ADO.NET provider,
however, can be made mappable by applying a SpatialSchema to the table definition. In
this method, the MDAL engine calls the ADO.NET data provider whenever data is
requested by a user. This tends to be a slower method of accessing data. However, when
used in conjunction with caching, it performs well. See Using an ADO.NET Data Provider.

Data from Third-Party Web Services

MapInfo Data Access Library (MDAL) can integrate Web service XML or GML output into
the Catalog for use in a MDAL desktop or web application. Data can be brought into the
Catalog and converted to a MultiPolygon, LineString, Point, or other Geometry via the
MapInfo Data Access Library (MDAL) API. MDAL then turns the Geometry into a
FeatureCollection, and, in turn, saves it to a memTable or native TAB format.
This approach is appropriate also if you wish to make data available for use in MDAL, but
not necessarily for map display.

Data Readers, MemTables and Result Sets

The methods to access data return a data reader or result set. A data reader allows
access in a sequential manner and does not store copies of data. It retrieves the data
from the data source, except in the case where the data source is cached. Result sets are
collections of keys. These keys allow you access back to the original tables and do not
create copies of the data.
A MemTable also allows you to store data from various sources into one table. This table
type stores data in a combination of memory arrays and temporary disk storage. When
data is added, the MemTable makes a copy of the data and does not have a key or

Pitney Bowes 31 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

pointer back to the original table. These are useful for temporary layers for maps and
containers for return values of processes such as a geocoding or routing result.
MemTable access and map rendering performance is equivalent to native tables.
Result sets are a great tool when you need access to a defined set of rows and when you
need to get data from the source. If the source data may change during your session then
this method allows you to see the results if the data source supports concurrent access.
Since MemTables are copies of data they are a static set of data rows and will not reflect
changes from the original data sources.

Using an ADO.NET Data Provider

Data that cannot be directly accessed with a specific TableInfo data source can use
TableInfoAdoNet. The ADO.NET table can be in one of two forms: DataTable (a collection
of rows from a single table kept in-memory and allows read-write access); or
IDbCommand (an SQL statement executed at the data source that yields read only,
dynamic data).

Accessing Data in a DataTable

When using a DataTable, the Catalog is essentially holding on to a reference to the
DataTable you supply to the call to Catalog.OpenTable (using the TableInfoAdoNet class).
DataTables are editable using the MapInfo ADO.NET Data Provider by issuing Insert,
Update, and/or Delete commands. Your application may continue to access the
DataTable directly as well. Note, however, that the structure of the table should not be
changed while the Catalog has a reference to it. Also note that changes to the data
outside of the MapInfo Data Provider (e.g., without using the MICommand to issue Insert,
Update, or Delete commands) will not result in the raising of the insert, update, or delete
table events.
The DataTable contains almost enough information for the Catalog to define the table. For
string columns, however, the Catalog needs to assign a length to this field. The length
would be used when constructing temporary indices, temporary tables for aggregation,
etc. For these types of operations, it is important to get the string length correct. The
DataColumn has a MaxLength property that should be set to indicate the maximum
length string the column could hold. If not set, this value defaults to -1 in which case the
value of 254 is used. Before checking the MaxLength property, the Catalog looks to see if
the DataColumn has a property defined in its ExtendedProperties collection with the
name “StringWidth”. If found, the value for this property is used as the column's width.
This example illustrates how to create a MapInfo Table whose data is stored in a
DataTable.

Pitney Bowes 32 Developer Guide

 2 – Working with Data

VB example:
Public Shared Sub MapInfo_Data_TableInfoAdoNet(ByVal connection As _
MIConnection)
' Create a new DataTable.
Dim dt As DataTable = New DataTable("CityData")
Dim dc As DataColumn
dc = dt.Columns.Add("City", Type.GetType("string"))
dc.MaxLength = 30
dc = dt.Columns.Add("Country", Type.GetType("string"))
dc.MaxLength = 30
dc = dt.Columns.Add("Continent", Type.GetType("string"))
dc.MaxLength = 30
dc = dt.Columns.Add("Population", Type.GetType("string"))

' Populate the DataTable...

dt.Rows.Add(New Object() {"Madrid", "Spain", "Europe", 1500000})
dt.Rows.Add(New Object() {"Stockholm", "Sweden", "Europe". 985000})

' Now open a MapInfo Table which accesses this DataTable

Dim ti As TableInfoAdoNet = New TableInfoAdoNet("Cities")
ti.ReadOnly = False
ti.DataTable = dt
Dim table As Table = connection.Catalog.OpenTable(ti)
End Sub

Accessing Data Using an IDbCommand

The second form of ADO.NET table is based on the connected object types in ADO.NET:
Connection, Command, and DataReader. MapInfo Tables constructed in this fashion are
read-only. These types of tables are created by passing to the Catalog an IDbCommand
object that is already configured to return all of the data that is to comprise the table.
When the table is initially created (by calling Catalog.OpenTable), ExecuteReader is
called on the IDbCommand. The resulting data reader is used to determine the columns
and their data types. All subsequent cursor requests (other than cursors which retrieve a
specific record - called a key fetch) also call ExecuteReader to fetch the data to satisfy
the cursor. Notice that this may be very inefficient. If at all possible, use one of the other
table types to access your data.
Since the Command-based form of the ADO.NET table is designed to use the generic
interfaces without requiring any specific knowledge of any particular implementation of
these interfaces, the table also does not assume that the IDbCommand.CommandText is
any form of standard SQL. In fact, it may not be SQL at all. This table type does not
access, parse, or modify the CommandText. This means that this table type has no
mechanism for knowing which column(s) in the results formulate a unique, non-null key

Pitney Bowes 33 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

value. For this type of table, it is required to tell the table which column(s) constitute the
key. This is accomplished by specifying the KeyType as Explicit and setting the
KeyColumns property.
There are many operations inside the MapInfo Data Provider which require the retrieval of
a specific record by key (also referred to as a key fetch). Select statements with a where
clause of the form MI_Key = '5' is a simple example in which we need to retrieve the
record whose MI_Key column can be represented by the string literal '5'. Key retrievals
are very common in mapping selections, labeling, and scrolling in a MIScrollableReader
(in which case the reader may be scrolling through a list of key values). MapInfo tables
are dependent upon the ability to efficiently fetch records by key value. Just as the
Command-based form of the ADO.NET table does not read, parse, or modify the
CommandText of the IDbCommand object that defines the table (the “Sequential”
IDbCommand), it has no ability to modify the IDbCommand object to fetch a specific
record. Thus, a second IDbCommand object must be supplied for this purpose. The
“FetchByKey” IDbCommand object must meet the following requirements:
• When ExecuteReader is called on this command object, it must produce a data reader
that has the same columns as the sequential command object and in the same order.
• The FetchByKeyCommand must contain a Parameters collection and must contain
one parameter for each member of the key. For example, if the TableInfo.KeyColumns
specifies a key as consisting of the “city” and “state” columns, then the
FetchByKeyCommand must contain two parameter objects. The first parameter object
is assigned a value representing the first column specified in the
TableInfo.KeyColumns collection (e.g., a value for “city”), the second parameter object
is assigned a value representing the second column specified in the
TableInfo.KeyColumns collection (e.g., a value for “state”), and so on. When
ExecuteReader is called on the FetchByKeyCommand, the reader must return the
record which represent s the specified key.
This example illustrates how to create a MapInfo Table that accesses data through the
ADO.NET connected command objects.
VB example:
Public Shared Sub MapInfo_Data_TableInfoAdoNet2(ByVal connection _
As MIConnection)
Dim ti As TableInfoAdoNet = New TableInfoAdoNet("EuropeanCities")
Dim _conn As OleDbConnection = New _
OleDbConnection("Provider=Microsoft.Jet.OLEDB.4.0;Data _
Source=C:\Data\EuropeCities.mdb")
Dim selectQuery As String = "SELECT City, Country, Continent, _
Pop_1994 FROM EuropeCities"
Dim _OleDbCommand As OleDbCommand = New OleDbCommand(selectQuery)
_OleDbCommand.Connection = _conn

Pitney Bowes 34 Developer Guide

 2 – Working with Data

selectQuery = selectQuery + " where City = @City AND _

Country = @Country"
Dim _OleDbKeyCommand As OleDbCommand = New _
OleDbCommand(selectQuery)
_OleDbKeyCommand.Parameters.Add("@City", OleDbType.Char)
_OleDbKeyCommand.Parameters.Add("@Country", OleDbType.Char)
_OleDbKeyCommand.Connection = _conn

' The MapInfo Table will Open/Close the connection as necessary.

' If this is expensive the application could open the connection
' before opening the table and closing the connection after the
' table is closed.
ti.SequentialCommand = _OleDbCommand
ti.FetchByKeyCommand = _OleDbKeyCommand

' Tell the table which column(s) constitute a key - for this table
' it is acompound key consisting of values from the City and County
' columns.
Dim sc As StringCollection = New StringCollection
sc.Add("City")
sc.Add("Country")
ti.KeyColumns = sc
ti.KeyType = KeyType.Explicit

' Ask the Catalog to open the table.

Dim tbl As Table = connection.Catalog.OpenTable(ti)

' Now the MICommand object may be used to select data from the table
' (by the name EuropeanCities since that is the alias we assigned to
' it). The data in thistable may be joined with any other table and
' it may be used as source data in a call to AddColumns to populate
' temporary columns with data from this table.
End Sub

Data Binding
Data binding is the process of bringing data from a data source into MapInfo Data Access
Library (MDAL). Data binding of external data (ADO.NET and other legacy sources) to
MapInfo.Data.Table is accomplished by opening an ADO.NET DataTable as a Table using
TableInfoAdoNet. The table can then be joined with another table, joined to itself or use
Table.AddColumns to bind columns to a second table.
To join a table to itself, following this example:
Select ... From T as A, T as B Where A.X = B.Y

If an application has data stored in a DataTable or data that is accessible through an

ADO.NET data provider, that data can be presented to the Catalog and treated as a
MapInfo table. This would be primarily useful if the data were not accessible through one
of the other table types.

Pitney Bowes 35 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

For example, if the data is stored in a dBase file, Microsoft Access table, or is accessible
through ODBC or Oracle's OCI interface, it is recommended that those TableInfo types be
used to access the data. Data which cannot be accessed through one of these types of
tables, but that can be loaded into a DataTable or is accessible through some ADO.NET
Data Provider that implements the Command, Parameter and DataReader object types
can still be accessed by the Catalog.
An application may need to make data available as a MapInfo native table so that queries
can be executed to join the data with other MapInfo table data. It may also need to be
made available to the Catalog so it can be used as the source data in a call to the
Table.AddColumns.

View Tables
A view is a way to relate information from one or more tables based on a named select
statement. The Catalog allows you to create views based on any table definition. View
tables have the following characteristics:
• The data is not copied.
• Access to views always accesses its base tables.
• View is an MapInfo SQL Select Statement with a name (Alias).
• Queries may be joins (forms composite keys).
• Membership in the View is live.
• Exception: Views that aggregate cache the data. Data changed events trigger
recomputation.
For more information and code examples, see the MapInfo.Data.TableInfoView class in
the Developer Reference Help system.

Result Sets

ResultSets are similar to view tables in that both are defined using a MapInfo SQL select
statement and have an associated name (Alias). ResultSets, however, have a fixed
membership of records based on the evaluation of the where clause (if any) at the time
the result set is created. Any access to the data in a ResultSet always reflects the data in
the source table. However changes to the source data will not cause the ResultSet to
add/remove a record based on the original where clause. ResultSets manage a set of
keys internally.
In general ResultSets are lightweight and temporary. Some of the characteristics of result
sets are:
• The data is not copied.

Pitney Bowes 36 Developer Guide

 2 – Working with Data

• Access to result sets always accesses its base tables.

• A ResultSet is a sorted list of keys, a collection of column definitions, and a name.
• Membership in the ResultSet is fixed.
• Exception: ResultSets that aggregate, cache the data. Data changed events trigger
recomputation.
• ResultSets are vulnerable to Delete and Pack operations.
For more information, see the MapInfo.Data.TableInfoResultSet class in the Developer
Reference Help system.

Source Rows
Source rows represent a match between the table records involved in Table.AddColumns.
When adding temporary columns to a table, multiple records from the data source may be
aggregated together to compute a value for each record in the destination table (also
referred to as the bind table). The MapInfo.Data.SourceRows class is a collection of
SourceRows that identify the records from the data source that were aggregated
together,
SourceRows only exist if the BindType property is DynamicCopy, which indicates that
changes to the source data are propagated to the temporary column automatically.

 Table.AddColumns is not supported for the following table types: Server, View,
Seamless, AdoNet, ResultSet, or Drilldown.

See also Adding Expression Columns to a Table.

The GeoDictionary
The GeoDictionary maintains information about which map entities can be matched to
which information. The GeoDictionaries class is a collection of GeoDictionary objects.
The MapInfo.Data.GeoDictionary namespace provides support for data autobinding by
being a programmatic representation of the GeoDictionary file. The GeoDictionary file
contains information about tables (TAB files only). The GeoDictionary is used to
automatically determine the table to which application data should be bound. The
GeoDictionary is persisted in a file (typically GeoDict.DCT) and is maintained using the
GeoDictionaryManager utility application.

Pitney Bowes 37 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

AutoMatching Using the GeoDictionary

The MatchResolver.AutoMatch method in the Data.GeoDictionary namespace initiates
the AutoMatching process. It does not call AddColumns, i.e., does not do the binding. A
subsequent call to BindColumn is required to perform the autobinding, or a direct call to
AutoMatchAndBind.
Automatching can encounter ambiguous situations. These situations include:
• multiple source columns are detected in the user data
• multiple tables/layers are detected that match the source column
• multiple geosets/workspaces are available for the matched table/layer.
It is the MatchResolver object with which the GeoDictionary communicates during the
match process to solve the ambiguity. It provides the matching algorithm. The basic class
selects the first or the one with the highest matching percentage. This class is not sealed
and client applications may derive their own class from this and override its behavior.

Making Tables Mappable

Tables can either be mappable (contain a GeometryColumn) or non-mappable (no spatial
attribute data). Mappable tables are added to a MapInfo Data Access Library (MDAL)
application as a layer in a map. Non-mappable tables, such as customer data, can be
made mappable when a GeometryColumn is created for it. MapInfo Data Access Library
(MDAL) provides spatial schemas to accomplish this.
Spatial schemas are services that can be applied to a table to enhance its spatial
capabilities. There are two type of spatial schemas: XY and PointRef. Non-mappable
tables that have attribute columns that represent X and Y values (such as longitude and
latitude) use SpatialSchemaXY and tables that have an attribute column which can be
used to reference a record in a mappable table uses SpatialSchemaPointRef.

SpatialSchemaXY
SpatialSchemaXY uses the X and Y values of each record in the table to construct point
objects and store them in a temporary column known as MI_Geometry. This spatial
schema may be applied to tables of any data source except Seamless, View, and
ResultSet.
By having a GeometryColumn, the table can now be displayed as a layer in a Map and
used for spatial analysis.
SpatialSchemaXY has the following characteristics:
• The Geometry column is editable.

Pitney Bowes 38 Developer Guide

 2 – Working with Data

• Editing the Geometry automatically changes the X and Y values.

• You can define styles for each point in the table.
• You can store the spatial information as a TAB file and open like any other table.
This spatial schema can be used for traditional server XY data without a MapCatalog.
(Using a MapCatalog may offer better performance on RDBMS's, since more work is
done on the server. See The MapInfo_MapCatalog.)
MI_Geometry is a temporary column unless you write out the TAB file explicitly using the
TableInfo.WriteToTab method. The schema is automatically regenerated when the table is
opened.
VB example:
Public Shared Sub MapInfo_Data_SpatialSchemaXY()
Dim ti As TableInfo = _
TableInfo.CreateFromFile("c:\data\customers.TAB")
' a non-mappable table
Dim xy As SpatialSchemaXY = New SpatialSchemaXY
xy.XColumn = "Xcoord"
xy.YColumn = "Ycoord"
xy.NullPoint = "0.0, 0.0"
' Any customer at 0,0 means we don't know their location.
xy.StyleType = StyleType.None
xy.CoordSys = _
Session.Current.CoordSysFactory.CreateLongLat(DatumID.WGS84)
ti.SpatialSchema = xy
' Now set the spatial schema information before
' opening the table.
Dim table As Table = Session.Current.Catalog.OpenTable(ti)
End Sub

Public Shared Sub MapInfo_Data_TableInfoNative2(ByVal ti As _

TableInfoNative)
ti.WriteTabFile()
End Sub

SpatialSchemaPointRef
This spatial schema uses a value in the table's data to create a Point geometry object by
matching the value against an equivalent value in a mappable table.
For example, if your table of customers contains addresses with postal codes, the
customer records can be tied to the spatial points in a postal code reference table.
SpatialSchemaPointRef is actually a join between two tables, one containing data and the
other containing a join column and an object column. The join column contains the same
values as the data column in the non-mappable table, such as postal codes. The result of

Pitney Bowes 39 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

applying SpatialSchemaPointRef is a table that contains a spatial geometry column for

records that were previously non-spatial. This geometry column has the following
characteristics:
• The data table may match more than one record in the geometry table. When this
happens the similar rows are aggregated into a MultiPoint geometry.
• The geometry is the centroid of the geometry from the other table.
SpatialSchemaPointRef has these characteristics:
• The temporary Geometry column is read-only.
• Any edits to a value in the reference table changes the Geometry value in the data
table.
• SpatialSchemaPointRef can be applied to any data source except Seamless, View,
and ResultSet.
• You can define styles for each point in the table.
• You can store table information as a TAB file and open like any other table.
For more information and code examples, see the MapInfo.Data.SpatialSchemaPointRef
class in the Developer Reference Help system.
VB example:
Public Shared Sub MapInfo_Data_SpatialSchemaPointRef(ByVal _
map As _Map)

' a non-mappable table

Dim ti As TableInfo = _
TableInfo.CreateFromFile("c:\data\customers.TAB")
Dim pr As SpatialSchemaPointRef = New SpatialSchemaPointRef
pr.CoordSys = map.GetDisplayCoordSys()
pr.StyleType = StyleType.None
pr.RefTable = "us_zips"

' the column in RefTable which will match the MatchColumn in my data
pr.RefColumn = "zipcode"

' a column in the Customer table

pr.MatchColumn = "zip"
pr.RefTableLocation = "c:\data\us_zips.tab"

' Now set the spatial schema information before opening the table.
ti.SpatialSchema = pr
Dim table As Table = Session.Current.Catalog.OpenTable(ti)
End Sub

Pitney Bowes 40 Developer Guide

 2 – Working with Data

MapInfo ADO.NET Data Provider

MapInfo Data Access Library (MDAL) provides mechanisms for issuing SQL commands
which return record sets from tables using ADO.NET. The MapInfo ADO.NET Data
Provider is one mechanism for accessing data in .NET applications in this fashion. For an
alternative that uses the Feature class and SearchInfo methods on the Catalog, see
Features and Feature Collections.

The following sections present the key interfaces and classes for accessing data via the
MapInfo ADO.NET Data Provider.
• MIConnection
• MICommand
• MIDataReader
• MapInfo SQL

MIConnection
An MIConnection represents a connection to the Catalog. The connection provides a
starting point for issuing SQL commands and obtaining results. Whereas most data
provider connections allow the user to immediately begin issuing queries or other
commands against existing tables (or schema objects), the MapInfo ADO.NET Data
Provider initially has no tables available. Tables need to be opened or created before they
can be accessed. When opened, a name (alias) can be associated with the table which is
used when resolving identifiers in the query engine.
Connections are not pooled in the MapInfo Data Provider and there is no connection
string required to create a new connection.

Pitney Bowes 41 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

The MapInfo.Engine.Session class creates and initializes the Catalog which may be
accessed through the Session.Current.Catalog property. The MIConnection.Open
method obtains a reference to the Catalog using the Session.Current.Catalog property
and the MIConnection.Close method sets the internal reference to the Catalog to null.
VB example:
Public Shared Sub MapInfo_Data_MIConnection()
Dim connection As MIConnection = New MIConnection
Dim command As MICommand = connection.CreateCommand()
command.CommandText = "Select * From States Where Pop > 1000000"

connection.Open()
Dim reader As MIDataReader = command.ExecuteReader()
Dim i As Integer, n As Integer = reader.FieldCount
For i = 0 To n - 1 Step i + 1
Console.Out.Write("{0}\t", reader.GetName(i))
Next
Console.Out.WriteLine()
While reader.Read()
For i = 0 To n - 1 Step i + 1
Dim o As Object = reader.GetValue(i)
If o Is DBNull.Value Then
Console.Write("null\t")
Else
Console.Write("{0}\t", o.ToString())
End If
Next
Console.Out.WriteLine()
End While
reader.Close()
command.Dispose()
connection.Close()
End Sub

MICommand
MICommand provides the necessary interface for executing SQL commands against the
MapInfo Data Provider. MICommand creates MIDataReader and MIScrollableReader
instances for obtaining data via the ExecuteReader and ExecuteScrollableReader
methods, respectively.

Supported Commands
The commands that are understood by the MICommand are:

Select
SELECT < select_list >
FROM { < table_source > } [ ,...n ]

Pitney Bowes 42 Developer Guide

 2 – Working with Data

[ WHERE < search_condition > ]

[ GROUP BY expression [ ,...n ] ]
[ ORDER BY {expression | column_position [ ASC | DESC ] } [ ,...n ]]

< select_list > ::=

{
*
| { table_name | table_alias }.*
| { expression } [ [ AS ] column_alias ]
} [ ,...n ]

< table_source > ::=

table_name [ [ AS ] table_alias ]

Insert
INSERT [INTO] { table_name } [ ( column_list ) ]
{ VALUES ({expression | NULL}[, ...n]) | query_specification

Update
UPDATE { table_name }
SET {{ column_name } = { expression | NULL }} [, ...n]
[WHERE < search_condition > ]

Delete
DELETE [FROM] { table_name } [ WHERE < search_condition > ]

< search_condition > ::=

{ [ NOT ] < predicate > | ( < search_condition > ) }
[ { AND | OR } [ NOT ] { < predicate > |
( < search_condition > ) } [ ,...n ] ]

< predicate > ::=

{
expression [ { = | < > | != | > | >= | < | <= } expression ]
| string_expression [ NOT ] LIKE string_expression [ ESCAPE
'escape_character' ]
| expression [ NOT ] BETWEEN expression AND expression
| expression IS [ NOT ] NULL
}

expression
Is a column name, pseudo column, column alias, constant, function, or any combination
of column names, column aliases, constants, and functions connected by an operator(s).
Column names and pseudo columns may be prefixed with a table name or a table alias
followed by the dot (“.”) character.
group_by_expression
Is a reference to a column in the select list - either an exact copy of the select list
expression, the alias, a 1-based number indicating the position of the column, or coln
where n is a number representing a column.

Pitney Bowes 43 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

order_by_expression
Is a reference to a column in the select list - either an exact copy of the select list
expression, the alias, a 1-based number indicating the position of the column, or coln
where n is a number representing a column.
For more information on expressions, where they are used and how to create them, see
Features and Feature Collections.

ExecuteFeatureCollection
The ExecuteFeatureCollection method in the MICommand class is the bridge between
the MapInfo ADO.NET Data Provider and the Feature object model. This method
executes command text (SQL statements) against the data source connection, and builds
an IResultSetFeatureCollection. The Feature model is discussed in Features and Feature
Collections.

MIDataReader
The MIDataReader provides forward-only, read-only access to the data returned from
executing a SQL Select statement. To create a MIDataReader, you must call the
ExecuteReader method of the MICommand object, rather than directly using a
constructor.
The MapInfo Data Provider allows multiple MIDataReader instances to be in use on the
same connection. However, if the Table being accessed resides on a Microsoft SQL
Server database, only one MIDataReader may be open at a time.
IsClosed and RecordsAffected are the only properties that you can call after the
MIDataReader is closed.Although the RecordsAffected property may be accessed while
the MIDataReader exists, always call Close before returning the value of
RecordsAffected to ensure an accurate return value.
You must explicitly call the Close method when you are through using the MIDataReader.
When accessing the DataReader through the IEnumerator or IFeatureEnumerator
interface, Close() is automatically called when MoveNext() returns false. Only one
enumerator can be used on a DataReader.

 For optimal performance, MIDataReader tries to avoid creating unnecessary

objects or making unnecessary copies of data. As a result, multiple calls to methods
such as GetValue may return a reference to the same object. Use caution if you are
modifying the underlying value of the objects returned by methods such as
GetValue.

Pitney Bowes 44 Developer Guide

 2 – Working with Data

The MIDataReader provides a means of reading a forward-only stream of rows from the
MapInfo data provider. This cursor type is the best performing for accessing a selection of
rows since there is little setup or overhead.

Scrollable Data Readers

MIScrollableReader derives from MIDataReader and offers forward and reverse reading.
Some of the options available with MIScrollableReader include:
• ReadPrevious
• Rewind
• Unwind
• ReadTop
• ReadBottom
• AtTop / AtBottom

 An MIScrollableReader is more expensive to create than MIDataReader. This is the

most expensive cursor since there is setup and extra resources necessary to keep
track of record order to allow scrolling. Use this cursor only if you need to scroll
through the record set.

MapInfo SQL
The MapInfo SQL Language allows you to add powerful analytical processing to your
MapInfo Data Access Library (MDAL) application. MapInfo Data Access Library (MDAL)
exposes SQL processing to users via the MapInfo ADO.NET Data Provider for accessing
data (specifically the MICommand object). Expressions are also used for labeling,
thematics, legends, AddColumns, Feature searching, and Selection processing.
MapInfo SQL is standardized based on SQL-3 Specification. For example, you will find
that:
• String constants are enclosed in single quotation marks
• Identifiers may be enclosed in double quotation marks
• Select has no relationship to the Selection
A complete reference including code examples for the MapInfo SQL language is provided
in the MapInfo SQL Reference, which you can view directly from Visual Studio’s Help
system.

Pitney Bowes 45 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Features and Feature Collections

The Feature class object model in MapInfo Data Access Library (MDAL) offers a non-
SQL-based approach to access and manipulate data. This section covers the Feature
class and IFeatureCollection interface. A key task in working with features is the ability to
search for them using a query definition object.

Feature
Features are described by their geometry, style, data source, key and attributes. Typically
a feature is a row in a table. A feature’s geometry is a FeatureGeometry object.
FeatureGeometries can cover a given area (MultiPolygon), a location (Points,
MultiPoints); and distance (MultiCurves, LegacyArcs). Additional Geometry classes that
derive from FeatureGeometry and are used for map features are
FeatureGeometryCollection and LegacyText. (Rectangle, rounded rectangle and ellipse
objects also derive from FeatureGeometry, but are used primarily for cosmetic display
purposes.)
One of the main uses of computerized maps is to gather information about the features.
In MapInfo Data Access Library (MDAL) features are returned in FeatureCollections by
any of several methods, either created from scratch using a schema, selected using
selection tools or methods or by searching the Catalog for those that meet a specific set
of criteria.
You can force a Load using the Load method. Changes made to the Feature are not
reflected in the underlying table (if there is one) until the Feature is saved back to the
table. This is done using the Update method, or UpdateFeature or InsertFeature. You can
throw away any edits done to the Feature object before it is saved using the DiscardEdits
method.

Pitney Bowes 46 Developer Guide

 2 – Working with Data

A Feature has a schema that describes the attributes of the Feature. The Columns
property describes the schema.

Retrieving Features from a Table

A Table is a type of Feature collection. As such, the Features within the table may be
enumerated directly. For example:
VB example:
Dim ftr As Feature
For Each ftr In table
...

The default feature enumerator for a table uses an MIDataReader internally with the
following command:
command.CommandText = "Select MI_Key, * From \"" + table.Alias + "\"";

To retrieve a subset of the features in a table, use one of the Catalog.Search methods or
use one of the MICommand.ExecuteFeatureCollection methods.

Modifying Features in a Table

To modify features in a table, use one of the following methods.
• Feature.Update
• Table.UpdateFeature
• Table.InsertFeature

Feature Collections
Feature collections are a group of Feature objects. All Features in a collection share the
same Schema (columns). The Feature collection has a schema which is the schema of all
of its member feature instances. Some Feature collections own their Features while other
Feature collections maintain references to Features.

Searching for Features

One of the most common tasks in Pitney Bowes’s mapping applications is to search for
features that meet certain criteria. Once you have the features you are interested in, you
can carry out further analysis, such as thematic mapping. In MapInfo Data Access Library
(MDAL), searching for features can be done in a number of ways: using tools, using
Catalog search methods, or using SQL and the MapInfo ADO.NET Data Provider.

Pitney Bowes 47 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

The following code sample shows two ways to search for the same thing, in this case,
cities in New York.
// Using SQL
command.CommandText = "Select Obj From States Where state = ‘NY’;
FeatureGeometry nyGeom = command.ExecuteScalar() as FeatureGeometry;
command.CommandText =
"SELECT * FROM Cities WHERE Obj within @newyork";
command.Parameters.Add("@newyork", nyGeom);
MIDataReader reader = command.ExecuteReader();
// or… to get a FeatureCollection
IFeatureCollection fc = command.ExecuteFeatureCollection();

// Using Features
Feature fNY = catalog.SearchForFeature("States", _
SearchInfoFactory.SearchWhere("state='NY'"));
SearchInfo si = SearchInfoFactory.SearchWithinFeature(fNY, _
ContainsFilter.ContainsType.Centroid);
IDynamicFeatureCollection dfc = _
catalog.Search("Cities", si) as IDynamicFeatureCollection;
Console.Out.WriteLine( _
"There are {0} cities whose centroid is within NewYork." _
dfc.Count);

SQL searches are more fully discussed in MapInfo ADO.NET Data Provider. The
following sections focus on searches using the Catalog and SearchInfo.

Catalog Search Methods

The Catalog has a number of search methods as members. The overloaded Search
method can be used to search on one or more tables. They include different arguments to
make each search unique. For example, the basic Search (Table, SearchInfo) searches
the given table and returns a FeatureCollection. The Search (ITableEnumerator,
SearchInfo) method searches on multiple tables and returns a
MultiResultSetFeatureCollection.
The SearchForFeature method returns the first Feature from the results. The
SearchReader method returns an MIDataReader cursor with the results.

Code Sample: SearchForFeature

The following example shows how to use Catalog.SearchForFeature and
Catalog.SearchWithinGeometry. It finds all the cities in the uscty_1k table that are within
Florida. It assumes that tables “usa” and “uscty_1k” are open and that there is one map.
VB example:
Public Shared Sub MapInfo_Data_SearchInfo(ByVal catalog As Catalog)

Pitney Bowes 48 Developer Guide

 2 – Working with Data

Dim fFlorida As Feature = _

catalog.SearchForFeature("usa",MapInfo.Data._
SearchInfoFactory.SearchWhere_("State='FL'"))
Dim si As SearchInfo =
MapInfo.Data.SearchInfoFactory.SearchWithinGeomeTry(fFlorida._
Geometry,ContainsType.Centroid)
Dim fc As IResultSetFeatureCollection = _
MapInfo.Engine.Session.Current.Catalog.Search("uscty_1k",si)

' Add results to selection.

MapInfo.Engine.Session.Current.Selections.DefaultSelection.Add(fc)
End Sub

SearchInfo and SearchInfoFactory

The MapInfo.Data.SearchInfo class defines the query used in a search and handles any
necessary post processing of the search results.
The SearchInfoFactory creates SearchInfo objects. SearchInfoFactory contains a number
of search methods that allow you to search using spatial references to your search
location or by using geometries that are drawn on the screen.
The following table describes the SearchInfoFactory search methods.

SearchInfoFactory
Methods Behavior

SearchAll Returns all the rows.

SearchNearest Returns the rows with table geometries that are closest
to the given search point.

SearchWhere Returns the rows specified by the given where Clause.

SearchWithinDistance Returns the rows where the table geometry is

contained within a distance of the search point,
rectangle or geometry. This method uses the
Geometry.Distance method to determine if an object is
in or out the search area. Previously
SearchWithinDistance had buffered the distance and
searched within the buffer, leading to less accurate
results.

SearchWithinFeature Returns the rows where the table geometry is

contained within the search features's geometry.

Pitney Bowes 49 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

SearchInfoFactory
Methods Behavior

SearchWithinGeometr Returns the rows where the table geometry is

y contained within the search geometry.

SearchWithinRect Returns the rows where the table geometry intersects

the given rectangle.

SearchIntersectsFeatu Returns the rows where the table geometry intersects

re with the search features's geometry.

SearchIntersectsGeo Returns the rows where the table geometry intersects

metry with the search geometry.

Analyzing Data
Once your data is available in the Catalog, you will want to analyze it to meet your
business objectives. The Catalog has an SQL processor that allows you parse and
aggregate your data. Here you have two options:
• OGC object-based query interface
• ADO.NET SQL-based interface
The diagram below shows the relationship between the two.
Group 1 shows the OGC query interface. Use these objects to construct a query. The
interface allows you to create queries to filter columns and rows, as well as add spatial
and non-spatial conditions. The queries interact through the Search methods off those
query objects to return data readers and result sets. Use these objects if you are more
comfortable with object-oriented programming and less so with SQL syntax. See
SearchInfo and SearchInfoFactory.
The ADO.NET interfaces, shown in group 2, use the defined ADO.NET model to allow
access via the MapInfo SQL language. The ADO.NET interfaces use SQL syntax to
interact with the Catalog. In this instance you need to generate the SQL statement and
assign it to the MICommand object. These objects use the Execute command to return a
data reader or result set. See MapInfo ADO.NET Data Provider.

Pitney Bowes 50 Developer Guide

 2 – Working with Data

MapInfo Data Access Library (MDAL) Data Model

Both the OGC query-based and ADO.NET command-based approaches use the Catalog
(group 3) to organize the data sources as a response to the object or SQL query. The
object-based query API will generate SQL and pass this to the Catalog for processing. In
some instances you may be able to generate more efficient SQL by hand, but the objects
are well defined and the interfaces restrict how you interact so the SQL tends to be
optimal. If you are comfortable with the SQL language using the ADO.NET method may
be more comfortable. But if you are inexperienced with SQL then the OGC object based
query will work just as well.
The MapInfo SQL syntax is defined in the SQL Reference which ships with MDAL. The
language is based on SQL3 and has special MapInfo operators defined for spatial
analysis. These operators begin with the MI_ prefix.

Data Readers, MemTables and Result Sets

The methods to access data return a data reader or result set. A data reader allows
access in a sequential manner and does not store copies of data. It retrieves the data
from the data source, except in the case where the data source is cached. Result sets are
collections of keys. These keys allow you access back to the original tables and do not
create copies of the data.

Pitney Bowes 51 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

A MemTable also allows you to store data from various sources into one table. This table
type stores data in a combination of memory arrays and temporary disk storage. When
data is added, the MemTable makes a copy of the data and does not have a key or
pointer back to the original table. These are useful for temporary layers for maps and
containers for return values of processes such as a geocoding or routing result.
MemTable access and map rendering performance is equivalent to native tables.
Result sets are a great tool when you need access to a defined set of rows and when you
need to get data from the source. If the source data may change during your session then
this method allows you to see the results if the data source supports concurrent access.
Since MemTables are copies of data they are a static set of data rows and will not reflect
changes from the original data sources.

Improving Data Access Performance

Performance is always an important aspect to any application that accesses data.
Consider the following list in your design and development plans for your application.
• Only request the data you need (especially from an RDBMS). This limits the amount of
data sent over the connection.
• Only sort tables if you need an ordered list. This process takes time to read through
the entire table to build an order. Also it will be slower if there is no index on the
column.
• Only scroll if you need random access to a table. This also builds indexes to speed up
access and remember order. Data readers access the data directly with no need to
read extra data.
• Use consistent coordinate systems for Join and Search operations. This eliminates the
need to convert geometries for every access.
• Use indexed columns for Join / Filter / Sort / Aggregate operations.
• Use CentroidWithin, ContainCentroid, and EnvelopesIntersect prior to actually
checking for geometry intersects. These tests are very quick and in most cases
eliminate a lot of geometries from your list with little effort.
• Use BeginAccess/EndAccess (especially for file-based tables) when performing
multiple queries and/or edits.
• Try to avoid calls such as Area and Buffer in the Where clause because the operation
will have to be done each time a new cursor is created.
• Try to avoid calls such as Area and Buffer in the Select list when defining a view or
result set for similar reasons.

Pitney Bowes 52 Developer Guide

 2 – Working with Data

• Use result sets for intermediate results or operations where you manage keys. These
are very light weight and afford quick direct access back to the original data.

Pitney Bowes 53 Developer Guide

3 – Working with Core MDAL
3
Classes

The MapInfo.Engine namespace contains the interfaces and classes

that relate directly to the core MDAL functionality. This includes the
core ISession interface which is the starting point for all MDAL
applications. Classes in this namespace include Session and
Selections, and SearchPath. Other types in the namespace are
supporting classes, delegates, structures, and enumerations for
Collections, Resources, and CustomProperties.

In this chapter:
 Session Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
 Selection Class . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
 Selection Code Examples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
 Event Arguments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
 Exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
 3 – Working with Core MDAL Classes

Session Interface
The ISession interface is the starting point for all applications integrating MDAL. It
manages the initialization of resources needed for a MDAL application.
An instance of ISession holds components of the object model such as the DataAccess
engine, MapFactory, CoordSysFactory so that the application can do work. The following
diagram illustrates the classes that implement ISession interface.

For an ASP.NET application each client request has its own ISession instance. This
instance resides in the calling context and is available throughout the lifetime of the
client's request.
For a single-threaded desktop application there is only one instance. On a multi-threaded
desktop application there is one instance per thread.
The MapInfo.Engine.Session class provides access to the ISession object. To get the
current ISession instance, access the MapInfo.Engine.Session.Current property.

Using Session.Dispose Method

The MapInfo.Engine.Session class has two overloaded Dispose methods. Your choice
will depend on the type of application you are building.

Pitney Bowes 55 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Session.Dispose()
Session.Dispose() disposes the ISession instance that is accessible through the
Session.Current property. This method is used only for multi-threaded desktop
applications.
Do not use this for single-threaded desktop applications. For single-threaded desktop
application, Dispose is called automatically when the application is shutdown or when the
AppDomain using MDAL is unloaded.

Selection Class
A Selection is a collection of IResultSetFeatureCollection objects that holds lists of
features. These features are a subset of rows in a table. They could be property
boundaries, street networks, cell tower locations, or natural features such as rivers. They
are typically drawn with special highlighting when they display in a Map. There can only
be one IResultSetFeatureCollection for a given table in a Selection.
There can be more than one Selection in a ISession. The Selections collection contains
all of the selections in the application. There is always at least one selection, known as
the DefaultSelection.
Each Selection must have a name and unique alias. By default, map selection tools
modify the Selection when used. Each tool can be set to use any particular Selection.
A selection in MDAL is not a copy; it is a reference to an IResultSetFeatureCollection for
a given table in a Selection. If you attempt to modify a Selection after you have closed the
table that you are working with, the reference to the IResultSetFeatureCollection will be
invalid, causing an exception.
Features can also be selected through search methods from the MapInfo.Data.Catalog
class which returns IResultSetFeatureCollection collections. A Selection object can be
passed into a search, which can be used to populate or change a Selection.
Features can also be selected via the ExecuteFeatureCollection method from the
Data.MICommand class. In this case, you would execute SQL commands against the
MapInfo Data Provider.
For more information on features, tables, the Catalog, and the MICommand see Working
with Data.

Pitney Bowes 56 Developer Guide

 3 – Working with Core MDAL Classes

SelectionChangedEvent
A delegate method is attached to the SelectionChangedEvent in order to receive
notification that this selection has changed. For example, if a record is added, the
SelectionChangedEvent is fired.

Selection Code Examples

The following are code examples of common selection operations. Additional code
examples are included in many topics of the MDAL Developer Reference.

Selecting Features Within Another Feature

A common search technique using MDAL is to find features within another feature. You
may do this to find all the customers within a postal code boundary or all the highways
that are under construction in a sector. Follow the example below. The parameter f is a
MapInfo.Data.Feature.
VB example:
Dim si As MapInfo.Data.SearchInfo = _
MapInfo.Data.SearchInfoFactory.SearchWithinFeature(f, _
MapInfo.Data.ContainsType.Centroid)
Dim irfc As MapInfo.Data.IResultSetFeatureCollection = _
MapInfo.Engine.Session.Current.Catalog.Search("USCty_8k", si)

MapInfo.Engine.Session.Current.Selections.DefaultSelection.Clear()
MapInfo.Engine.Session.Current.Selections.DefaultSelection.Add(irfc)

irfc.Close()

Checking a Table for Selections

Follow the code example below to learn how to get a count of selections in a table.
VB example:
Public Shared Sub MapInfo_Engine_Selection2()
Dim session As ISession = MapInfo.Engine.Session.Current
Dim tableUsa As Table = session.Catalog("usa")

' Get fc for selection on usa.

Dim fc As IResultSetFeatureCollection = _
session.Selections.DefaultSelection(tableUsa)
Dim nCount As Integer = 0
If Not fc Is Nothing Then

Pitney Bowes 57 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

nCount = fc.Count
End If
End Sub

You can also perform selection operations using MapInfo SQL queries and with the
ADO.NET data provider. See Chapter 2 Working with Data.

Returning All Columns From a Table

The following sample shows how to return all columns from a selection:
VB example:
Dim Connection As MIConnection = New MIConnection
Connection.Open()
Dim ti As MapInfo.Data.Table = _
MapInfo.Engine.Session.Current.Catalog.GetTable("usa")
Dim si As MapInfo.Data.SearchInfo = _
MapInfo.Data.SearchInfoFactory.SearchAll()
si.QueryDefinition.SetColumns("*")
Dim irfc As MapInfo.Data.IResultSetFeatureCollection = _
MapInfo.Engine.Session.Current.Catalog.Search(ti.Alias, si)
Dim l As MapInfo.Data.Feature
For Each l In irfc
Dim column As MapInfo.Data.Column
For Each column In l.Columns
MessageBox.Show(column.ToString())
Next
Next

Event Arguments
The MapInfo.Engine namespace contains various event argument classes that provide
data for events. Refer to the online help for more information. Some of the event
argument classes include:
• CollectionCancelableEventArgs – Provides data for a collection event that can be
cancelled.
• CollectionEventArgs – Provides data for a collection event.
• NodeSelectionChangedEventArgs – Fires these event arguments when the node
selection changes.
• SelectionChangedEventArgs – Other objects can attach delegates to this event to get
notified when a selection changes.

Pitney Bowes 58 Developer Guide

 3 – Working with Core MDAL Classes

Exceptions
The Engine namespace contains various exception classes. Refer to the online help for
more information. Some of the exception classes include:
• ResourceNotFoundException – Throws this type of exception when the requested
object is not found in the Resource table.
• ResourceTypeMismatchException – The exception that is thrown when the object
read from Resources was not of the expected type.
• TimeoutException – The exception is thrown on Current timeout while waiting for a
pooled ISession to become available.

Pitney Bowes 59 Developer Guide

4 – Creating Expressions
4
Expressions are used throughout MDAL to describe the exact pieces
of information you need to display and analyze in your mapping
application. This chapter covers creating expressions for a wide range
of product areas, including data access, creating themes, labeling
maps and more.

In this chapter:
 Expressions Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
 Creating Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
 Where Clause – Boolean Expressions . . . . . . . . . . . . . . . . . . . . . . . 62
 Functions In Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
 Expression Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
 4 – Creating Expressions

Expressions Overview
Expressions are statements that are used to describe and format data. For example, in
English, an expression might read like “a median income of more than $50,000, or
“female percent of population.”
Expressions are formed using column names, constants (i.e., specific data values), along
with functions and operators that act upon the columns and constants. The operators and
functions are defined in the MapInfo SQL Language, developed to support MapInfo .NET
supported products. For details, see the MapInfo SQL Reference via the Help Viewer in
Visual Studio.
Use expressions to make the most of your data. By using expressions you can:
• Show only the columns and rows of data that interest you.
• Derive new columns by calculating new values based on the contents of your existing
columns.
• Aggregate data to work with subtotals instead of the entire table.
• Combine data from two or more tables into one results table.
Many of the data sets you will use include more objects and information than necessary
for your projects. In many cases it is easier to work with a subset of the complete data
product. For example, if you were tracking crime statistics for a certain county by census
tract, you would not need the census tracts for the entire state. You would use an
expression to extract just the census tracts for the county.
Expressions are used throughout MDAL, in the following areas:
• SQL statements (select, insert, update, delete, group by, order by)
• SQL functions that take expressions as an arguments (e.g., the geometry argument in
MI_Area() is an expression that returns a geometry object.)
• Adding columns (MapInfo.Data.Table.AddColumn creates a temporary column based
on an expression.)
• Feature searches (SearchInfo and SearchInfoFactory)

Creating Expressions
The simplest possible expression consists of a constant, such as “2” (numeric example)
or “Friday” (text example).
Other simple expressions consist of a column name, for example:

Pitney Bowes 61 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

POP_2000
STATE

When you request specific multiple columns in a select statement, for example, these
columns together are known as an expression list.
Select colA, colB, colC from Table1, Table2
Select colA/2, ColB/ColC from Table1

You can also write expressions that perform mathematical operations on your data.
For example, RENT + UTILITIES is an expression that adds two columns together. This
expression could be used in a SQL statement to find all apartments that have a total cost
of less than $800 per month.

Where Clause – Boolean Expressions

A Boolean expression is a search condition that results in a value of either True or False.
For example, the expression
2 < 5

is a Boolean expression because the result is True.

All expressions that contain relational operators, such as the less than sign (<), are
Boolean. The operators AND, OR, and NOT, are Boolean operators. Boolean
expressions are also called comparison expressions, conditional expressions, and
relational expressions.
POP_2000 > 500000
POP_2000 <= POP_1990
PROVINCE <> ‘Ontario’
County = ‘Columbia’ AND VALUE >= 250000

Supported operators in MapInfo SQL are defined in the MapInfo SQL Reference.
Boolean expressions are used in the “where clause” of an SQL statement. The where
clause is the expression that controls the rows that are returned (the rows that result in
True).
For example, the boolean expression in this statement follows WHERE. Only objects in
the Europe table that fall within the boundary of France will be returned as True.
"SELECT * FROM Europe WHERE MI_Geometry within @France";

Functions In Expressions

Pitney Bowes 62 Developer Guide

 4 – Creating Expressions

Functions in MapInfo SQL are used to create even more complex expressions to retrieve
data that meets specific criteria. For example, MapInfo SQL supports many of the usual
database functions that work with strings, dates/time, and numbers,
The most powerful functions in MapInfo SQL are those that take advantage of the spatial
nature of mapping data. These geographic functions are used to create new geometries,
measure area and length, return spatial information, validate spatial relationships among
geometries, and others. Supported functions are defined in the MapInfo SQL reference.
An example of using a function in an expression might be when you wish to look at the
area of a table of boundaries, such as school districts. Use the function MI_Area() to
return the area of each record in the table.
Additional examples of functions in expressions are found in the Expressions Examples
section below.

DateTime and Time Expressions

When using DateTime and Time Expressions with MDAL, please be aware of the
following:
• If a DateTime column or Time column is used alone in an expression, it is formatted
using the current locale.
• If a DateTime or Time column is in an expression, its string value is TimeToNumber or
DateToNumber + space + TimetoNumber.
• Operator math on Time or DateTime is not supported. You can add a number to a
Date, but not to a Time or DateTime.

Expression Examples
The following highlights some uses of expressions.

SQL Statement Examples

This example will select all records from the Eurcity_1K table that are within Germany
and have a population of over 1 million.
Select * from Eurcity_1K WHERE (MI_Geometry MI_Within @Germany) AND Tot_Pop >
1000000

The following examples make selections based on Time and Date columns in the table.
This example will select all crime records from a "CrimeActivity" table where the crime
occurred between 12:00:00 AM and 6:00:00 AM:

Pitney Bowes 63 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

SELECT * FROM CrimeActivity WHERE CrimeTime BETWEEN '12:00:00 AM' AND '6:00:00
AM'

Where CrimeTime is a Time column that stores the time at which the crime occurred.
This example will select employee Names from an "Employee" table who were born
before December 31, 1970.
SELECT Names FROM Employee WHERE BirthDay < '12/31/1970 12:00:00 AM'

Where BirthDay is a Date column that stores the birthdays of employees.

MapInfo SQL Function Example

The following expression uses a MapInfo SQL function to find features within a buffer.
Obj CentroidWithin MI_Buffer(Obj, 5, ‘km’, ‘Spherical’, 24)

This expression uses a MapInfo SQL special keyword reserved for geographic objects
called ‘Obj’. This keyword describes the geometry of the object such as its coordinate
system and bounds. This keyword is compatible with previous versions of MapX and
MapInfo Professional. It is equivalent to the column name MI_Geometry.
Note that km and Spherical are enclosed in single quotes. In MapInfo SQL, string literals
must be enclosed in single quotation marks while identifiers such as column names, table
names, aliases, etc.) should be enclosed in double quotation marks, but only needed if
the parsing logic is unable to correctly parse the identifier. This would include identifiers
that have spaces in their names or other special characters.
To find features that fall outside the buffer, the expression would look like:
NOT Obj CentroidWithin MI_Buffer(Obj, 5, ‘km’, ‘Spherical’, 24)

Add Columns Example

When adding temporary (computed) columns to a table using the AddColumns method,
the columns supplied contain an expression that defines how the value for the column is
computed. The expression may contain an aggregation function if multiple source records
are expected to match up to a single record in the table to which the columns are being
added.
The example below uses expressions to represent population density "Pop_1990 /
MI_Area(Obj, 'sq mi', 'Spherical')". The expressions are preceded by their new column
names. PopDensity1990 and PopDensity2000.
VB example:
Public Shared Sub MapInfo_Data_TableAddColumns(ByVal miTable _
As Table)
Dim NewCols As Columns = New Columns

Pitney Bowes 64 Developer Guide

 4 – Creating Expressions

NewCols.Add(New Column("PopDensity1990", "Pop_1990 / _

MI_Area(Obj, 'sq mi', 'Spherical')"))
NewCols.Add(New Column("PopDensity2000", "Pop_2000 / _
MI_Area(Obj, 'sq mi', 'Spherical')"))
miTable.AddColumns(NewCols)
End Sub

For more information on adding columns, see Adding Expression Columns to a Table.

Feature Search Example

The following example uses a boolean expression SearchWhere("State='FL'") that, when
executed, will return a value of 1 for each row that contains FL.
VB example:
Public Shared Sub MapInfo_Data_SearchInfo(ByVal catalog As Catalog)
Dim fFlorida As Feature = catalog.SearchForFeature("usa", _
MapInfo.Data.SearchInfoFactory.SearchWhere("State='FL'"))
Dim si As SearchInfo = _
MapInfo.Data.SearchInfoFactory.SearchWithinGeometry(fFlorida._
Geometry, ContainsType.Centroid)
Dim fc As IResultSetFeatureCollection = _
MapInfo.Engine.Session.Current.Catalog.Search("uscty_1k", si)

' Set the map view to show search results

MapInfo.Engine.Session.Current.MapFactory(0).SetView(fc.Envelope)
' Set the view of the first map.

' Add results to selection.

MapInfo.Engine.Session.Current.Selections.DefaultSelection.Add(fc)
End Sub

For more information on the Feature class and search methods, see Features and
Feature Collections.

Pitney Bowes 65 Developer Guide

5 – Accessing Data from a
5
DBMS

MapInfo Data Access Library (MDAL) provides access to a number of

spatially aware DBMS. This is a powerful feature that allows
developers to connect to live data stored in spatial servers, such as
Microsoft SQL Server or the Oracle Spatial databases. Spatial servers
allow companies to host their map data in their enterprise database for
central management and security.

In this chapter:
 Accessing Remote Spatial Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
 Accessing Remote Tables Through a .TAB File . . . . . . . . . . . . . . . . 67
 Accessing Remote Tables Without a .TAB File. . . . . . . . . . . . . . . . . 67
 Mapping DBMS Data with X/Y Columns. . . . . . . . . . . . . . . . . . . . . . 68
 Accessing Data from Oracle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
 Accessing Data from MS SQL Server . . . . . . . . . . . . . . . . . . . . . . . 72
 DBMS Connection String Format . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
 Defining Mappable Tables in Server Table Queries . . . . . . . . . . . . . 77
 Accessing Attribute Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
 Performance Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
 Working with the Cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
 The MapInfo_MapCatalog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
 Adding Rows to the MapInfo_MapCatalog . . . . . . . . . . . . . . . . . . . . 86
 Per-Record Styles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
 5 – Accessing Data from a DBMS

Accessing Remote Spatial Data

You can access data using MDAL with different DBMS servers. The servers include:
• Microsoft Access 2007 and Excel 2007
• Microsoft Access 2003
• Oracle 11G (11.1.0.6.0 and 11.1.0.7.0)
• Oracle 10G, 10GR2
• Microsoft SQL Server 2012 (with SQL Native Client 11)
• Microsoft SQL Server 2008 (with SQL Native Client 10)
• Microsoft SQL Server 2014
You can add a table from data in a DBMS using the TableInfoServer class in the
MapInfo.Data namespace.
The details for adding spatial data are included in the following sections.

Accessing Remote Tables Through a .TAB File

Using the MDAL, an application can access DBMS data “live”, or can open a MapInfo
Professional linked table. However, the linked table will be read-only and cannot be
refreshed by your application. The data is actually from the remote database and does
not reflect the data in the local linked version.
You can create a .TAB file to provide access to remote data. To generate a .TAB file using
MapInfo Professional, choose File > Open a DBMS table.
The .TAB file is a text file; you can create a .tab file using any text editor. Once you have
created the .tab file, you can access it the way you access any other MapInfo .TAB file
programmatically through the Catalog object or through the Workspace Manager.

Accessing Remote Tables Without a .TAB File

An application does not need a .TAB file to access remote data. The following code
sample illustrates this process.
VB example:
Public Shared Sub MapInfo_Data_TableInfoServer(ByVal connection As _
MIConnection)
' Note: Do not specify any columns. These are determined

Pitney Bowes 67 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

' dynamically from the query

Dim ti As TableInfoServer = New TableInfoServer("Provinces")

ti.ConnectString = "SRVR=ontario;UID=mapx;PWD=mapx"
ti.Query = "Select * From Provinces"
ti.Toolkit = ServerToolkit.Oci

ti.CacheSettings.CacheType = CacheOption.Off ' On is the default

Dim tbl As Table = connection.Catalog.OpenTable(ti)

End Sub

Mapping DBMS Data with X/Y Columns

You can access data from a DBMS table that has X/Y coordinates. You need to create a
MapInfo_MapCatalog and register the tables as SpatialType 4.0 and specify two column
names as the coordinates. The columns should be indexed on the table. Connect to the
DBMS via ODBC and specify the new columns as “Obj” or “MI_Geometry” in your query.

Accessing Data from Oracle

To connect to an Oracle database from an MDAL application, the Oracle OCI connectivity
client must be installed and appropriate permissions granted. See your Oracle
documentation for detailed information.

Geometry Conversion
The table below shows the translation from MDAL objects to Oracle Spatial
(SDO_GEOMETRY).

From MapInfo To Oracle

NULL geometry NULL

Point 1 POINT

MultiCurve (with IsLegacyLine 2 LINESTRING Geometry contains one line

= true) string

Polygon 3 POLYGON Geometry contains one polygon.

Pitney Bowes 68 Developer Guide

 5 – Accessing Data from a DBMS

From MapInfo To Oracle

FeatureGeometryCollection 4 Collection Geometry is a heterogeneous

collection of elements.

MultiPoint 5 MULTIPOINT

MultiCurve 6 MULTILINESTRING Geometry has multiple

line strings.

MultiPolygon 7 MULTIPOLYGON Geometry has multiple

polygons.

Ellipse NULL

LegacyArc NULL

Rectangle NULL

LegacyText NULL

RoundedRectangle NULL

PieTheme, BarTheme NULL

The table below describes the translation from Oracle (GTYPES) to MapInfo Spatial
objects.

From Oracle GTYPES To MapInfo

0 *UNKNOWN_GEOMETRY (Spatial ignores this

geometry.)

1 POINT Geometry contains one point. Point

2 LINESTRING Geometry contains one line string. MultiCurve

3 POLYGON Geometry contains one polygon. MultiPolygon

4 *Collection Geometry is a heterogeneous collection FeatureGeometryColle

of elements. ction

5 MULTIPOINT Geometry has multiple points. MultiPoint

Pitney Bowes 69 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

From Oracle GTYPES To MapInfo

6 MULTILINESTRING Geometry has multiple line MultiCurve

strings.

7 MULTIPOLYGON Geometry has multiple polygons MultiPolygon

(more than one exterior boundary).

*Some data loss may occur when translating to or from MapInfo object format.
They are interpreted (when possible) as single point SDO_POINTTYPE values if
not already NULL. They “grab” the first point in the ordered array which would be
interpreted as a NULL geometry.

Oracle Support for Z and M Values

MDAL supports reading and writing Oracle GTYPEs with Z and M values. The presence
and order of Z and M is determined by inspecting the DIM_INFO array in the
USER_SDO_GEOM_METADATA for the table. MDAL checks for the following dimension
names (case insensitive):
• For Z dimension: "Z", "Elevation", " Depth" and "Z Dimension"
• For M dimension: “M", "Measure", and "M Dimension"
Tables that contain M and/or Z values now return FeatureGeometry objects that contain
the data for the dimensions present. FeatureGeometry instances inserted or updated into
an Oracle table will preserve each of the four dimensions of the new geometry (XYZM)
that the Oracle table is defined to support. For geometries containing dimensions
unsupported by the Oracle table, the values for those dimensions are ignored during
insert/update operations. For geometries not containing dimensions supported by the
Oracle table, NULL values are supplied for the missing dimensions during insert/update
operations. For example, when inserting a geometry with no Z or M values into a table
that is defined with dimensions x, y, and m, the M values stored in the table will be NULL.
The success of any of these insert/update operations may also be dependent upon
additional server-side validation including explicit column constraints and validation of
values against the dimensional extents specified in the SDO_GEOM_METADATA system
table.

SDO_GEOMETRY Arc and Circle Translation

Circles and circular arcs can be resolved to MultiCurves with a resolution of 25 segments
per 360 degree circle.

Pitney Bowes 70 Developer Guide

 5 – Accessing Data from a DBMS

Visualization of Non-translatable Oracle Objects

An Oracle Spatial Object that your MDAL application is unable to translate produces a
Point object with a default style (a black star) at the location of the SDO_Spatial point, or
the first SDO_Spatial ordinate in the ordinate array. This is to enable a visual
representation of the non-translatable object in the proper geographic area to which it
belongs. Examples of non-translatable objects are user-defined objects GTypes 0,4,5, or
invalid SDO_geometries containing unrecognized GTypes, ETypes, or interpretations.
The second class should also fail using SDO_VALIDATE_GEOMETRY().

Centroid Support
An MDAL application uses the SDO_POINT as the centroid value for polygons. This
centroid feature is used to position labels, and also affects the tool selection of the object.
The Oracle SDO_GEOMETRY.SDO_POINT_TYPE field (if not NULL) is interpreted as
the feature centroid if the point exists inside the region. If the point exists outside of the
region, its centroid is calculated as always.

 There is currently no method or tool in MDAL to set the centroid of a region feature,
but one may read and use a stored centroid.

Oracle Spatial Reference Support (SRID)

An Oracle SDO_GEOMETRY column may be defined with a spatial referencing system.
This is done by providing the Oracle SRID in the USER_SDO_GEOM_METADATA and
also by assigning that SRID in the stored SDO_GEOMETRY values. If a table contains an
Oracle Spatial column with an assigned SRID, your MDAL application is able to query
and properly interpret the data. The MapInfo_MapCatalog must contain the same
MapInfo Professional CoordSys string as indicated in the SRID of the data, since it is the
Coordsys in the MapInfo_MapCatalog that is currently used to interpret and update the
data.
If the Spatial column does not contain an SRID value, (the value is NULL), your MDAL
application is also able to interpret the data via the MapInfo Professional Coordsys
defined in the MapCatalog.
When loading tables that use the Latitude/Longitude coordinate system (Geodetic Data)
to Oracle Spatial, it is important to verify that all geometry coordinates are between (-
180,180) longitude and (-90, 90) latitude. Geodetic data coordinates beyond that range

Pitney Bowes 71 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

are not supported in Oracle Spatial and may cause problems. You can check your data
using MapInfo Professional before loading, or use the Oracle Spatial
SDO_GEOM.VALIDATE_LAYER() function on the table after loading it to Oracle Spatial.

Accessing Data from MS SQL Server

MDAL supports data stored in Microsoft’s SQL Server 2008, SQL Server 2012 and SQL
Server 2014. The following information pertains to SQL Server 2008.

SQL Server 2008 Support

MDAL supports reading and writing data from and to Microsoft SQL Server 2008,
including the spatial data types GEOMETRY and GEOGRAPHY, along with M and Z
value support for both spatial formats.
To access data from SQL Server 2008, MDAL requires SQL Server Native Client 10. Data
is then handled like data from other remote database management systems that MDAL
supports1. Use the MapInfo.Data.TableInfoServer class to define the connection string
and an SQL statement to execute on the remote table. Internally, MDAL uses ODBC to
access the remote database.
The following table shows how objects are handled in MDAL given a specific object type
from SQL Server 2008.

SQL Server 2008 Spatial

GEOGRAPHY or GEOMETRY MDAL FeatureGeometry

Sql Server 2008 Spatial FeatureGeometry

GEOGRAPHY/GEOMETRY

Point Point

LineString MultiCurve

Polygon Multipolygon

MultiPoint MultiPoint

MultiLineString MultiCurve

MultiPolygon MultiPolygon

1.

Pitney Bowes 72 Developer Guide

 5 – Accessing Data from a DBMS

SQL Server 2008 Spatial

GEOGRAPHY or GEOMETRY MDAL FeatureGeometry

GeometryCollection FeatureGeometryCollection

GeometryCollection containing MultiPoint

only Points and/or MultiPoints

Geometrycollection containing only MultiCurve

LineStrings and/or MultiLineString

Geometrycollection containing only MultiPolygon

Polygons and/or MultiPolygons

An EMPTY NULL
GEOMETRY/GEOGRAPHY, e.g.,
Point empty

This table shows how an MDAL FeatureGeometry is written back to SQL Server 2008

SQL Server 2008 Spatial

MDAL FeatureGeometry GEOGRAPHY or GEOMETRY

Point Point

MultiPoint MultiPoint

MultiPoint containing only one Point

Point

MultiCurve MultiLineString

MultiCurve containing only one LineString

Curve comprised of two points

Multipolygon MultiPolygon

FeatureGeometryCollection GeometryCollection *

Rectangle NULL

RoundedRectangle NULL

Ellipse NULL

Pitney Bowes 73 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

SQL Server 2008 Spatial

MDAL FeatureGeometry GEOGRAPHY or GEOMETRY

LegacyArc NULL

LegacyText NULL
* This GeometryCollection may contain any or all of the following types: MultiPoint,
MultiLineString, and MultiPolygon.

SQL Server 2008 provides new types for date and time information. The following table
shows how date and time types are mapped to MDAL date and time types.

SQL Server MDAL

Date Date

Time Time

DateTime DateTime

SmallDateTime DateTime

DateTime2 DateTime

DateTimeOffset No support

Spatial tables from SQL Server 2008 must be registered in the MapInfo_MapCatalog so
that MDAL understands what it reads.
The MapCatalog provides four new spatialcolumn values to represent SQL Server 2008
tables:
17.x for GEOMETRY without M and Z values
18.x for GEOGRAPHY without M and Z values
20.x for GEOMETRY with M and Z values
21.x for GEOGRAPHY with M and Z values.
Data can be uploaded using MapInfo Professional or EasyLoader or you can use MapInfo
Professional to make existing data mappable, which will create the entry in the
MapCatalog. See The MapInfo_MapCatalog for more information on the MapCatalog.

Pitney Bowes 74 Developer Guide

 5 – Accessing Data from a DBMS

DBMS Connection String Format

ODBC Connection String Format
The format of the ODBC connection string is defined by several clauses separated by
semicolons (;). Each clause has the form Key=Value. Important keys are listed below.

Keyword Description

DSN= Specifies the ODBC data source name.

Caution: If you use the DSN= syntax key, the name that you specify
must match the data source name in use on the user’s system. Note
that different users might use different names to refer to the same
data source. If you cannot know in advance what data source name
to use, use the DRIVER= syntax key instead of the DSN= syntax
key.

DRIVER= Specifies the exact driver name of the installed driver. Used instead
of the DSN= syntax key.
Example:
DRIVER={SQL Server}

 This does not apply to Oracle Spatial.

UID= Specifies the desired UserId for the data source, if required.

PWD= Specifies the user’s password for the data source, if required.
Passwords do not need to be in the connection string for the two
strings to match. If two tables are in the same database, the
connection string is the same.

Oracle Spatial Connection String Format

These are the Oracle Spatial keywords. The string is defined by several clauses
separated by semicolons (;). Each clause has the form Key=Value. Important keys are
listed in the table below.

Pitney Bowes 75 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Keyword Description

SRVR= Reflects the service name for the server set in the Oracle Net8
EasyConfig utility. This is required for Oracle connectivity, but does
not apply to ODBC connections.

UID= Specifies the desired UserId for the data source, if required.

PWD= Specifies the user’s password for the data source, if required.

Sample Connection Strings

Here are sample connection strings for Oracle Spatial and Microsoft SQL Server 2008
ODBC drivers.

Oracle Spatial connection string:

UID=george;PWD=password;SRVR=OracleSpatial9i

Microsoft SQL Server 2008 connection string:

DRIVER={<driver>};
SERVER=<server>;UID=<uid>;PWD=<pwd>;Database=<database>

where <driver> for SQL Server Spatial should be the most current available, SQL Server
Native Client 10.0 or higher version.

Defining Mappable Tables in Server Table

Queries
The query you specify for a server table defines the result set of data from your DBMS
that represents the data in the table being added. You can formulate a fairly complex
query to do powerful server-side analysis that defines a mappable table in MDAL. Your
MDAL application uses this query internally to access the data.
MDAL generates several internal queries based on your query to access the data in a
map as well as selection/key based queries. The table from which the geometry column is
selected must be registered in the MapInfo MapCatalog on the server. MDAL requires this
to obtain certain metadata about the geometry column such as the coordinate system,
spatial type, and default styles.

Pitney Bowes 76 Developer Guide

 5 – Accessing Data from a DBMS

In order for a query to define a mappable table, the query must contain both a geometry
column and a key column. Sometimes for more complex queries on small sets of data
(where the spatial indexing or spatial predicate cause the query to fail), you can specify
TableInfoServer.MbrSearch=false to enable the results to be mapped.

The Geometry Column

If you do not specify a geometry column that your MDAL application can recognize, the
table is opened, but cannot be added to a map (the table is unmappable). MDAL
determines the geometry column of the table by looking it up in the MapCatalog and by
examining the result set datatype of the column. You can reference the geometry column
generically via the pseudo column name “Obj”, or you may refer to the geometry column
using its specific column name. This form is required to reference the geometry column
for an X/Y mappable layer. You can specify a geometry column via any server-supported
geometry function/expression.

Example
Select Obj from rdbsdata
Select sw_geometry from rdbsdata
select sw_member, ST_Buffer(geometry, 66.0, 0.1) from rdbsdata
// a geometry function
Select st_geometry(st_point(72.5, 42.5) from rdbsdata
// a geometry constructor

Oracle sdo_buffer example:

Select mi_prinx, mdsys.sdo_geom.sdo_buffer(geoloc, (select diminfo from
sdo_geom_metadata where table_name = 'ALINE'), 20) from aline where prinx = 1

Oracle constructor example:

Select 1 "mi_prinx",
mdsys.sdo_geometry(3,null,null,mdsys.sdo_elem_info_array(1,3,3),
mdsys.sdo_ordinate_array(-79.919909,40.553465,-71.060457,45.363657)) from dual

SQL Server 2008 Spatial function example:

select location_id, geography::Point(lat, long, 4326 /*WGS84*/) as geog from
dbo.store_locations

The Key Column(s)

A key column(s) must be returned in the query to enable it to be opened as a table. This
is what enables your MDAL application to identify each row in the result set.

Pitney Bowes 77 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

The key column does not need to be specified in the query in most cases.
Your MDAL application can look up and determine the best key column(s) to use in order
to uniquely reference a row in the result set, and then add them to the query if they are
not present. In most cases, this is the primary key/unique index.
For Oracle Spatial tables, the MI_PRINX may be used.
For some queries, it is not possible for your MDAL application to identify the key. This is
the case in a query on a view or a synonym. The view or synonym must appear in the
MapInfo MapCatalog. They also must be registered as required with the underlying
Spatial index system in most cases. Since MDAL cannot determine the key on these, a
mechanism is provided to allow the application developer/query writer to identify the key
column in the result set. The key must be a single column and must be a distinct value in
the result set. To identify the column that is to be used as the key column, you can specify
column alias of prinx or mi_prinx, (e.g., select custid mi_prinx, custname, Obj from
mycust).

Example
Select customer_id mi_prinx, obj from customer_view

The column alias “mi_prinx” is used to identify and use the customer_id column as the
key column for the table. You can alternately alias the desired key column in the create
view statement to identify the key column automatically for any query on that view.

Example
Create view customer_view as select customer_id mi_prinx, geoloc from customer

In general, if a column name or column alias of prinx, or mi_prinx is found in the result set,
that column is used as the key column for the table. This enables the application/query
writer to specify the key column they desire.

Accessing Attribute Data

To use all available data columns, specify a query such as Select * From tablename. You
are not required to specify * (asterisk); instead, you can designate specifically which
columns you want to use. For the best performance, limit your query so that it retrieves
only the needed columns.
When you add a DBMS table, for performance sake, you should only specify the columns
in the query that you intend to use in your application. These are the spatial column, the
key column(s), which are added automatically if you do not specify them, and columns
you want to label with, or create a theme from. You may use the pseudo columns “OBJ”

Pitney Bowes 78 Developer Guide

 5 – Accessing Data from a DBMS

for any mappable table to refer to the column(s) containing the spatial data. This is
required for a table using the MapMarker MDIGEOADDRESS column on a table with an
X/Y column.
You can use any server side expression/function to specify a column. Also, avoid select *
from tab in a real application.
The following code example defines a server table using a TableInfoServer.
VB example:
Public Shared Sub MapInfo_Data_TableInfoServer(ByVal connection As _
MIConnection)
' Note: Do not specify any columns. These are determined
' dynamically from the query

Dim ti As TableInfoServer = New TableInfoServer("Provinces")

ti.ConnectString = "SRVR=ontario;UID=mapx;PWD=mapx"
ti.Query = "Select * From Provinces"
ti.Toolkit = ServerToolkit.Oci

ti.CacheSettings.CacheType = CacheOption.Off ' On is the default

Dim tbl As Table = connection.Catalog.OpenTable(ti)
End Sub

Performance Issues
Establishing a connection with the database server may take several seconds. This is a
one-time cost, incurred when the table is first opened.
The retrieval speed depends on how much data is retrieved from the server. In some
cases, working with data from a server is noticeably slower than displaying a map from a
local file. Speed also depends on whether your MDAL application has already cached the
map features that are being fetched.

Working with the Cache

Knowing how to work with cache in MDAL enables you to improve your application’s
performance. The sections below describe what the cache is, how it works in the MDAL
object model, and the CacheSettings property of the TableInfoServer object.

What Is the Cache?

Pitney Bowes 79 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

In place of local files, applications can access MDAL table features from a remote
database. In place of reading these records from the database each time the map needs
to be acted upon, your application can temporarily store these records in the cache. This
limits the number of calls between the application and the remote database. Records in a
server table can be cached to improve the performance of your application. Server table
data is cached internally as it is read. All subsequent redraws read from this cache
instead of going to the server database for the same data. The cache is able to offer
significant redraw performance improvement.
There are several settings that developers can use to customize cache usage. The cache
can be enabled (or disabled) when the server table is added by specifying the values for
the CacheSettings property of the TableInfoServer object and is On by default.

How the Cache Works

For each record that is cached, each attribute data value is stored in memory and each
feature object is stored on disk in a temporary Rtree file. For tables with a lot of records
and/or a large record size (for example, number of bytes per record for the attribute data),
caching may use a significant amount of memory. If an application tries to cache too
much data, too much virtual memory usage may be required, which can degrade
performance. Applications should be selective about how the cache is utilized. MDAL
offers a variety of mechanisms for controlling the cache.

Pitney Bowes 80 Developer Guide

 5 – Accessing Data from a DBMS

The TableInfoServer Object and the CacheSettings Property

When a table is added to the map, the cache is enabled by default but can be further
configured using the CacheSettings property of the TableInfo object. This property has
four possible values: ON, OFF, ALL, USER, with ON being the default for
TableInfoServer, OFF is default for other TableInfo objects.

Parameter Description

OFF A value of 'Off' means that the table will not use the cache at all. All
data operations will go directly to the database server.

ON Caching is enabled and the table automatically performs caching of

the retrieved records. The user may additionally control the cache
through the cache constraint objects.
To avoid having a cache that grows excessively large, there are
controls that can be placed on the table's cache to determine when
to discard old cached data. These controls are properties of the
CacheParameters object, which can be set at the time the table is
initially opened. This allows the developer to set limits on the
maximum amount of memory or disk space used by the cache,
and/or the maximum number of records to maintain in the cache.
These limits can be used individually or in combination to provide
the cache management that best suits the application's needs.
ON is the default setting for the CacheParameter setting for a
TableInfoServer. For other TableInfo data sources, the default is
OFF. For example, TAB files are not cached by default.

Pitney Bowes 81 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Parameter Description

USER A value of USER for the LayerInfo CACHE parameter means that
your application creates a cache, but the only records that are
placed in the cache are those specified by the application
developer. The mechanisms available for specifying which records
are placed in the cache are BoundConstraint, FeaturesConstraint,
and AllFeaturesConstraint objects. The word constraint implies that
these objects are constraining the cache to include the specified
records. The BoundsConstraint object can be used to place all
records into the cache for which the MBR of the feature intersects
the MBR of the constraint.
A FeaturesConstraint object can be used to add specific records to
the cache. For example, if an analysis is going to be performed that
involves multiple steps and/or reads of the Feature or RowValues of
the feature, possibly on a set of features returned from a
Layer.Search, Layer.SearchWithinDistance, etc., it may be
advantageous to place these records into the local cache for the
duration of the analysis and remove them when finished. The
FeaturesConstraint provides this capability. If an application is
going to perform an analytically intensive operation that may hit
every record, it may be desirable to temporarily cache the entire set
of data for the layer. This is accomplished by using the
AllFeaturesConstraint. These cache constraint objects can also be
used when the cache is set to ON. In this case, they may add
records to the cache but have no effect on the cache's history of
previous map window views. The constraint objects can also be
used when the cache is set to OFF or ALL in which case they have
no effect.

 The constraint objects have no effect on non-server tables.

ALL The entire table is cached. With this option, the table's data is
retrieved from the server once and accessed locally from that point
forward. To refresh the data in the cache, use the Refresh method
on the table.

If you try to cache too much data or too many tables, virtual memory usage may be
forced, and performance gain could be lost.

Pitney Bowes 82 Developer Guide

 5 – Accessing Data from a DBMS

Cache Storage Type:

Cache Storage type indicates the table type used to store the cached records of RDB
server tables. Following Storage Types are used in MDAL:
1. Native: The cached records are stored on disk in a temporary MapInfo Native table. If
multiple tables are opened with this cache storage type, then a different set of MAP,
DAT, etc. files will be created for each table. This storage type has a maximum file-size
limits of 2GB.
2. MemTable: The cached records are stored in memory in a temporary MemTable table.
Geometry objects are stored on disk in a temporary MAP file. If multiple tables are
opened with this cache storage type, then a separate set of in-memory cache and on-
disk MAP files will be created for each table. This storage type also has an upper size
limit of 2GB.
3. NativeX: The cached records are stored on disk in a temporary MapInfo NativeX table.
If multiple tables are opened with this cache storage type, then a different set of MAP,
DAT, etc files will be created for each table. This NativeX storage type supports UTF-8
and UTF-16 charsets as well as it can store the data files that are greater than 2GB in
size.
4. MemNativeX: The cached records are stored in memory in a temporary MemNativeX
table. Geometry objects are stored on disk in a temporary MAP file. If multiple tables
are opened with this cache storage type, then a separate set of in-memory cache and
on-disk MAP files will be created for each table. MemNativeX cache supports UTF-8
and UTF-16 charsets as well as it can store the data files that are greater than 2GB in
size.
5. Geopackage: The cached records are stored on disk in a temporary Geopackage
database(*.GPKG) file. All RDB server tables opened in a session with this cache
storage type are cached in a single Geopackage database and for each open RDB
table, a separate Geopackage cache table is created in that single Geopackage
database.
6. MemGeopackage: The cached records are stored in an in-memory Geopackage
database.

The MapInfo_MapCatalog
In order to display data on a map, your MDAL application needs to access a special table,
known as the MapInfo_MapCatalog. One catalog must be created per database before
any tables in that database can be viewed as a map layer in an MDAL application. The

Pitney Bowes 83 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

MapCatalog must contain information about the spatial columns in each of the mappable
tables you want to access from the database. The MapInfo EasyLoader utility
automatically inserts the appropriate row into the MapInfo_MapCatalog when the table is
uploaded into the database.
Your application can use a MapInfo_MapCatalog that already exists on the server. (This
same catalog is shared by various MapInfo client applications). If there is no
MapInfo_MapCatalog on the server, you need to create one. MDAL supports the storage
of style information for individual features in remote databases.

Loading Spatial Data to DBMS

If you have spatial data in the form of a MapInfo table, you can import it into your DBMS
database.
To load data into Microsoft SQL Server and Oracle Spatial, use the MapInfo EasyLoader,
that is distributed with MapInfo Professional and available for download from
www.pbinsight.com. The EasyLoader utility automatically creates a MapInfo_MapCatalog
when you upload a table, if there is no MapInfo_MapCatalog already present.

Manually Creating a MapInfo MapCatalog

If you are not a MapInfo Professional or EasyLoader user, you or your database
administrator will need to create the MapCatalog manually, as described below. You only
have to create the MapCatalog once per server/database.
1. Create the user MAPINFO in the specific database where the mappable tables are
located.
2. Create the table MAPINFO_MAPCATALOG in the database.
The Create Table statement needs to be equivalent to the following SQL Create Table
statement:
Create Table MAPINFO_MAPCATALOG (
SPATIALTYPE Float,
TABLENAME Char(32),
OWNERNAME Char(32),
SPATIALCOLUMN Char(32),
DB_X_LL Float,
DB_Y_LL Float,
DB_X_UR Float,
DB_Y_UR Float,
VIEW_X_LL Float,
VIEW_Y_LL Float,
VIEW_X_UR Float,

Pitney Bowes 84 Developer Guide

 5 – Accessing Data from a DBMS

VIEW_Y_UR Float,
COORDINATESYSTEM Char(254),
SYMBOL Char(254),
XCOLUMNNAME Char(32),
YCOLUMNNAME Char(32),
RENDITIONTYPE INTEGER,
RENDITIONCOLUMN CHAR(32),
RENDITIONTABLE CHAR(32)
NUMBER_ROWS INTEGER
)

 It is important that the structure of the table looks exactly like this statement. The
only substitution that can be made is for databases that support varchar or text
data types; these data types can be substituted for the Char data type.

3. Create a unique index on the TABLENAME and the OWNERNAME, so only one table
for each owner can be made mappable.
create unique index mapcat_i1
on mapinfo.mapinfo_mapcatalog (OwnerName,TableName)

4. Grant Select, Update, Insert, and Delete privileges on the MAPINFO_MAPCATALOG.

This allows users to make tables mappable.
grant select, insert, update, delete on mapinfo.mapinfo_mapcatalog to public

Pitney Bowes 85 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Adding Rows to the MapInfo_MapCatalog

For each spatial table that you want to access from your application, you need to add a
row to the MAPINFO_MAPCATALOG table. If you do not use MapInfo Professional to
manage the MapInfo_MapCatalog, you will have to add rows to the
MAPINFO_MAPCATALOG table manually.
The following table describes the syntax and meaning of each column:

Column Name Values to Assign Examples

SPATIALTYPE MapInfo Spatial Object Format 14.0 = SQL Server

1: Point layer in X/Y columns indexed 14.1
with micode (a serialized quadtree key) 14.2
4: Point layer in X/Y columns 14.3
5.x: SpatialWare for Oracle
6.x: Ingres SQL - Not Supported
7.x: Sybase SQS - Not Supported
8.x: Oracle SDO version 2 - Not Supported
9.x: MapInfo Geocoding DataBlade
SpatialWare Point Module
10.x: MapInfo Geocoding DataBlade XY
Module
11.x: SpatialWare IDS/UDO datablade
13.x: Oracle Spatial
14.x: SpatialWare for Microsoft SQL Server
17.x:SQL Server 2008 GEOMETRY with-
out M and
Z values
18.X: SQL Server 2008 GEOGRAPHY with-
out M and Z values

Pitney Bowes 86 Developer Guide

 5 – Accessing Data from a DBMS

SPATIALTYPE 20.x: SQL Server 2008 GEOMETRY with M

(continued) and Z values
21.x: SQL Server 2008 GEOGRAPHY with M
and Z values
Spatial Object Type
x.0: Points only
x.1: Lines only
x.2: Regions only
x.3: All types supported

 This column describes the Spatial

Object Format of how the data is stored
and indexed and the Spatial Object
type(s) supported and not supported in
the column. The digits to the left of the
decimal point are the Spatial Object
Format. The digits to the right
represent the type of Spatial Object
Type that can be stored in the column.

Maps to
MapInfo.GeometryColumn.PredominantGe-
ometry Type, and
Has<Line/Point/Region/Text>Geometries

TABLENAME The name of the table. STATES

Pitney Bowes 87 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

SPATIALCOL- The name of the column, if any, containing SW_GEOMETRY

UMN spatial features:
SW_GEOMETRY (mappable using Spatial-
Ware Type/IDS/UDO)
NO_COLUMN (mappable using X–Y)
MI_SQL_MICODE (mappable using MI
Code) Or the name of the IDS/UDO, or Ora-
cle column that is ST_SPATIAL datatype.
Name of the Oracle SDO_GEOMETRY col-
umn.

DB_X_LL The X coordinate of the lower left corner of -360

the layer’s bounding rectangle, in units that
are indicated by the COORDINATESYS-
TEM (see below). Maps to Map-
Info.Data.GeometryColumn.Bounds

DB_Y_LL The lower left bounding Y value. -90

DB_X_UR The upper right bounding X value. 360

DB_Y_UR The upper right bounding Y value. 90

VIEW_X_LL The X coordinate of the lower left corner of -180

the default view. The default view only
applies if this is the first table to be opened.
Maps to
MapInfo.Data.GeometryColumn.Default-
View

VIEW_Y_LL The lower left bounding Y value of the -45

default view.

Pitney Bowes 88 Developer Guide

 5 – Accessing Data from a DBMS

VIEW_X_UR The upper right bounding X value of the 180

default view.

VIEW_Y_UR The upper right bounding Y value of the 45

default view.

COORDINATE- A string representing a MapInfo CoordSys Earth Projection 1,

SYSTEM clause (but without the keyword CoordSys at
0
the very start), which specifies a map projec-
tion, coordinate units, etc. For simple
Lon/Lat maps, specify Earth Projection 1, 0.
Maps to MapInfo.Data.GeometryCol-
umn.CoordSys

SYMBOL A MapInfo Symbol clause (if the layer con- Symbol(35,0,12)

tains only points); or a Symbol clause fol-
Pen(1,2,0)
lowed by a Pen clause (indicating styles for
linear features) followed by another Pen Pen(1,2,0)
clause (indicating styles for the borders of Brush(2,255,255)
regions) followed by a Brush clause. Maps to
MapInfo.Data.GeometryColumn.Default-
Style

XCOLUMN- For the X/Y mappable tables, specify the NO_COLUMN

NAME name of the column containing X-coordi-
nates. If there is no such column (i.e., if this
table uses a single spatial column instead of a
pair of X-Y columns) then specify NO_COL-
UMN or leave empty. Maps to Map-
Info.Data.SpatialSchemaXY

YCOLUMN- For the X/Y mappable tables, specify the NO_COLUMN

NAME name of the column containing Y–coordi-
nates, or specify
NO_COLUMN Maps to MapInfo.Data.Spa-
tialSchemaXY

Pitney Bowes 89 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

RENDITION- This indicates how the object style informa- 0 or 1

TYPE tion is applied.
0 – Indicates that all the objects in the table
will have the style specified in the symbol
field of the MapCatalog applied to them. No
per-record styles are in effect. Objects will be
read/updated using the default style for the
table.
1 – Indicates that the table uses per-record
styles. The table has a separate column that
contains a MapBasic string representation of
the style information for each object in the
table (the same format that is currently used
in the MapCatalog’s SYMBOL column). The
style column in the table is recorded in REN-
DITIONCOLUMN.

RENDITION- If RENDITIONTYPE is 1, this field stores MI_SYMBOLOGY

COLUMN the name of the column in the spatial table
that contains style information. This column
is automatically added to any query against
the table and is maintained (updated) as the
object is updated. Users should NOT specify
this column in their queries as problems can
occur with intersect or update statements.
Queries which include this column in the
select clause (excluding the wildcard charac-
ter “*”) may access the values through the
Dataset object. Rows with a NULL value in
their style column will have the style from the
SYMBOL field of the MapCatalog applied to
the object. Creates a MapInfo.Column.Data-
Type with MIDBType.Style

Pitney Bowes 90 Developer Guide

 5 – Accessing Data from a DBMS

Currently not used, but reserved for future Null

use.

NUM- Currently used by MapInfo Professional. Null

BER_ROWS

Per-Record Styles
Per-record style support brings a feature to spatial database implementations that has
long been available in MapInfo TAB files. Specifically, it allows each geometry in a single
table to have its own style. For example, a single 'public institution' table in Oracle Spatial
can have schools, town halls, libraries, and police departments and each point type would
be represented with its own symbol (i.e., a school symbol for all the schools). Similarly, a
single road table in SpatialWare SQL Server may have different road types such that
streets are shown as a single pixel black line, secondary roads as a double pixel red line
and interstates as parallel red lines.
To use per-record styles, your table must be represented with an entry in the MapCatalog
with appropriate settings for RENDITIONTYPE, RENDITIONCOLUMN, and
RENDITIONTABLE.

 If these columns are not present, the table’s default style will be applied to all
objects.

Symbol, Pen, Brush Clause Syntax

If you are manually creating a MAPINFO_MAPCATALOG table to provide support for a
remote spatial database, you will need to specify a symbol style, and possibly line and fill
styles as well.

Pitney Bowes 91 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Specifying Point Styles

Use a Symbol clause to specify point styles. There are three types of Symbol clauses:
one for specifying MapInfo 3.0-style symbols; one for specifying TrueType font symbols;
and one for specifying bitmap symbols.

Symbol Syntax Example

Symbol(shape, color, size) Symbol(35,0,12)

or
Symbol(64,255,12,"MapInfo
Symbol(shape,color,size,font,fontstyle,r Weather",17,0)
otation) Symbol("sign.bmp", 255, 18, 0)
or
Symbol(bitmapname,color,size,custom
style)

Specifying Line Styles

Use a Pen clause to specify line styles. In a MapInfo_MapCatalog, you may need to
specify two pen clauses: one to specify the appearance of linear features, and another to
specify the appearance of region borders.

Pen Syntax Example

Pen(thickness, pattern, color) Pen(1, 2, 0)

Specifying Fill Styles

Use a Brush clause to specify the style for closed features (regions)..

Brush Syntax Example

Brush(pattern,color,backgroundcolor) Brush(2, 255, 65535)

Text Objects Limitation

LegacyText objects have their own way of displaying style that is separate from the use of
the MI_Style column. Therefore any form of text object needs to be treated differently
than other objects. The style for any text object is embedded and a NULL value is
inserted into the style column.

Pitney Bowes 92 Developer Guide

 5 – Accessing Data from a DBMS

Troubleshooting
If you encounter problems with your SpatialWare or Oracle applications, use the following
table to help analyze and solve the problem.

Problem Description Possible Cause Solution

The table is not Data binding was AddColumns is not

matchable. attempted against a currently supported for
SpatialWare layer. SpatialWare layers.

No object was found A query was made Check that the table name
using the index that you against a table that does is correct and in the proper
specified. not exist. case. Also, the table may
need to be mappable.

No spatial object is Use the EasyLoader

contained in the result of Upload utility to make the
the spatial query. table a mappable table.

A query was made Check the query for

against a non-spatial possible syntax errors.
table. Also make sure that the
result of the query includes
the field specified in the
spatial column in the
MapInfo_MapCatalog.

Map appears to have The MBR for a DBMS Edit the extents
incorrect zoom level. layer is determined by (DB_X_LL, DB_X_UR,
For example, the map the MapInfo_MapCatalog DB_Y_LL, DB_Y_UR) in
may be zoomed out too table. The table extents the MapInfo_MapCatalog
far to identify any in the MapCatalog result using the MapInfo
geography. in a different zoom level Professional MBX tool,
than the one you desire MISETMBR.MBX.
for your output.

Pitney Bowes 93 Developer Guide

6 – Spatial Objects and
6
Coordinate Systems

This chapter covers the MapInfo.Geometry namespace and provides

descriptions and examples for writing applications for creating and
manipulating geometry objects.

In this chapter:
 Introduction to MapInfo.Geometry Namespace . . . . . . . . . . . . . . . . 95
 Geometries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
 Checking for Points in Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
 Checking for Points in Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
 Coordinate Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
 6 – Spatial Objects and Coordinate Systems

Introduction to MapInfo.Geometry Namespace

The MapInfo.Geometry namespace is used for creating and manipulating geometry
objects, and the coordinate systems in which they are used. Geometry objects are used
in maps to represent single points, such as cities (represented as point objects),
boundary lines, such as county borders (represented by MultiCurve objects), and regions,
such as countries or zip code areas (represented by MultiPolygon objects).
The classes, interfaces, and enumerations in the MapInfo.Geometry namespace define
the types representing the geometries and coordinate systems used in displaying
geographic features on a map. The Geometry model provides support for Z and M values
on FeatureGeometry objects. Interfaces allow for creation and editing of the geometry
objects. Methods such as Buffer, Combine, Difference, and Intersection provide object
processing on single objects or pairs of objects.

Geometries
The Geometry class allows for the creation, editing, and other manipulation of geometry
objects. Classes which inherit from the Geometry class and represent types of Geometry
objects include Point, MultiPoint, Polygon, MultiPolygon, Curve, CurveSegment,
LineString, and Ring. The following legacy classes are also inherited from the Geometry
class: Rectangle, RoundedRectangle, Ellipse, LegacyArc, and LegacyText.
The Geometry class represents the topmost level of the MapInfo Geometry object model.
This class is abstract, and cannot be instantiated. All classes that derive from this class
contain knowledge concerning their coordinate system. All classes are able to make
copies of themselves, and compare themselves to other Geometry objects for equality.
The diagram below shows a representation of the Geometry model.

Pitney Bowes 95 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Geometry Objects
All geometry objects in MDAL are created with a specific coordinate system that cannot
be changed. If you need to alter the coordinate system of an object you can make a copy
of that object in the new coordinate system.

Editing Geometry Objects

All Geometry objects contain a method for retrieving an interface to an editor that places
the object into Edit Mode. Once editing is finished the EditingComplete() method needs to
be called to signify that the editing of the object is complete. When the EditingComplete()
method is called, the order of the objects contained by the Geometry is reshuffled and all
references to them are dropped and need to be re-established in order to access them
again.
For example, the user creates a MultiPolygon and then edits the MultiPolygon. If the user
inadvertently moves a node of the interior ring to be outside of its containing Polygon the
Polygon is no longer valid. When EditComplete is called, all the contained objects within
the MultiPolygon are reshuffled, fixing the problem.
The geometry objects in the MDAL Object Model are described in the following sections.

Pitney Bowes 96 Developer Guide

 6 – Spatial Objects and Coordinate Systems

FeatureGeometry Objects
The FeatureGeometry class is specifically designed to contain classes that can be placed
in tables and that can be parts of Features and FeatureCollections. In order for something
to be displayed in a map, it needs to be in a table. FeatureGeometry objects are by
definition included in tables. Any object that is a subclass of Geometry and not a subclass
of FeatureGeometry cannot be saved to a table or included as part of a Feature or
FeatureCollection. An exception is thrown, or the program will not compile, if such an
operation is attempted. The FeatureGeometry class, like the Geometry class is abstract
and cannot be instantiated.

Support for M and Z Values

Feature geometries support reading and writing M and Z values at each node of the
object.
Support for reading and writing M and Z values for linear objects was accomplished by
extending the MDAL Geometry model. FeatureGeometry objects (Point, MultiPoint,
MultiPolygon, MultiCurve and FeatureGeometryCollection) can hold values for X, Y, Z
and M for each node.
IsMeasured and Is3D properties allow you to determine whether the object has M or Z
values. Additional properties and methods are provided to read and modify M or Z values
at each node. The minimum and maximum ranges of M and Z values can be retrieved as
well.
MDAL provides creation and editing capabilities for FeatureGeometries. For more
information, see MapInfo.Geometry.FeatureGeometry class in the Developer Reference.
M values for MultiCurves provide valuable information in linear network applications for
tracking and managing assets, events and conditions.

Point
Points are derived from the FeatureGeometry class and represent a single point on a
map. Points can be contained within a MultiPoint collection and then operated upon
collectively.
Use the following example code to model the creation of a Point:
using MapInfo.Geometry;

CoordSys longLatNad83;
CoordSysFactory coordSysFactory = new CoordSysFactory();
longLatNad83 = coordSysFactory.CreateLongLat
(MapInfo.Geometry.DatumID.NAD83);
DPoint point = new DPoint(0.0, 0.0);

Pitney Bowes 97 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Point pointGeometry = new Point(LongLatNad83, point);

MultiPoint
A MultiPoint contains an unordered, disconnected set of Points and is useful for
performing multiple operations on multiple points.
Use the following example code to model the creation of a MultiPoint object:
using MapInfo.Geometry;

CoordSys longLatNad83;
CoordSysFactory coordSysFactory = new CoordSysFactory();
longLatNad83=coordSysFactory.CreateLongLat
(MapInfo.Geometry.DatumID.NAD83);
MultiPoint multiPointGeometry = new MultiPoint
(longLatNad83, pointArray);

where pointArray is an array of DPoints.

MultiCurve
The MultiCurve class is derived from the FeatureGeometry class, and contains a possibly
disconnected set of Curves. These Curves may interact in many ways; they can be
connected or disconnected, and can intersect or overlap each other.
Although the Geometry object model supports multiple CurveSegments for each Curve,
the current version of the MapInfo engine is limited to having one CurveSegment per
Curve that is part of a FeatureGeometry (i.e., MultiCurve). This limitation derives from the
current TAB file format, which remains largely unchanged for this version of MapInfo.
Hence, the limitation concerns FeatureGeometry objects only.
Upon construction of a MultiCurve where the constructor takes a Curve or Curves which
may contain multiple CurveSegments per Curve, the actual Curves contained in the
constructed MultiCurve are altered to always contain only one CurveSegment per Curve.
Currently, the only types of CurveSegments that exist are LineStrings. Curves containing
multiple LineString CurveSegments have the LineStrings combined to form one large
LineString.
Upon completion of editing (signified by calling EditingComplete(), any Curve which was
added to the MultiCurve and contained multiple CurveSegments is altered in a similar
manner as noted above to produce Curves containing single CurveSegments.This
limitation, of Curves contained in MultiCurves always containing only a single
CurveSegment, should be removed in future versions of MapInfo as new types of
CurveSegments are introduced (e.g., EllipticalArcs, CircularArcs, and Splines), and the
TAB file format is altered. Also, during construction and on completion of editing, any
empty Curves are automatically removed from the MultiCurve.

Pitney Bowes 98 Developer Guide

 6 – Spatial Objects and Coordinate Systems

Line objects made up of two points that exist in MapInfo TAB files become MultiCurve
FeatureGeometry objects. They can be detected as two-point Lines by using the
IsLegacyLine property of the MultiCurve:
See the Developer Reference for a code example of creating and editing a MultiCurve
object.

Measure Values on MultiCurves

The Geometry object model supports M and Z values on FeatureGeometry objects. M, or

measure values, hold data at the nodes of MultiCurve objects that describes anything you
wish to map and analyze, including physical assets, conditions or events. The M values
play an important role in linear referencing and dynamic segmentation.

Curve Sort Order

The order of the Curves in a MultiCurve may be altered during construction, as compared
to the array of Curves passed to the constructor, and upon completion of editing. Due to
this, plus the removal of empty Curves, and limitations in the current implementation, any
references to Curves contained in a MultiCurve prior to and during editing may no longer
be valid after editing is completed (i.e., after EditingComplete() is called). If these objects
are referenced, they throw an ObjectDisposedException. After editing, the parts of a
FeatureGeometry should be reacquired to obtain a valid reference.

LineStrings
A LineString is a directed collection of sequential points that are connected in a linear
manner. Any two consecutive points in the LineString are connected by a straight line.
LineStrings can be part of Curves or Rings, or they can exist as a stand-alone Geometry.
LineStrings that are part of Curves or Rings inherit the coordinate system of their
container. Stand-alone LineStrings can be empty. A LineString that is contained in a
Curve or Ring that is not in Edit Mode cannot be empty, and must contain at least two
points.
See the Developer Reference for a code example.

Rectangle
A Rectangle Geometry contains two points representing the lower left hand and upper
right hand corners of the Rectangle. The other two points are implied. Rectangles are
always axis aligned, and always appear to be rectangular in shape, regardless of the
coordinate system, and are not projected. They do not contain any warping that may be
represented by the coordinate system.

Pitney Bowes 99 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

See the Developer Reference for a code example.

RoundedRectangle
A Rounded Rectangle behaves exactly like a Rectangle but is displayed with the corners
appearing rounded as a display-time only feature. The corners display as quarter circles
and the radius of the circle is controlled by the CornerRadius parameter.
Because RoundedRectangle objects, like rectangle objects, are defined by two points and
always display axis-aligned and unprojected, they are designed to be used primarily for
cosmetic display purposes. While many operations are available using Rectangle objects
(e.g., Combine), internally, a MultiPolygon copy of the Rectangle is used for these
operations. The resulting MultiPolygon contains 5 points (with the first and last points
being identical), and are effected by the coordinate system. In some instances, the
converted Rectangle may no longer appear rectangular. Use the CreateMultiPolygon
method to convert a RoundedRectangle to a FeatureGeometry object.
See the Developer Reference for a code example.

Ellipse
The Ellipse is inscribed in an axis-aligned rectangle defined by a DRect. The DRect is
defined by two points, the opposite corners of a rectangle, with the other two corners of
the rectangle implied. The Ellipse displays as if it were unprojected, regardless of the
coordinate system, and any skew that may be represented by the coordinate system.
Because Ellipse objects are defined by two points and always display axis-aligned and
unprojected, they are designed to be used primarily for cosmetic display purposes. While
many operations are available using Ellipse objects, internally, a MultiPolygon copy of the
Ellipse is used for these operations. The resulting MultiPolygon is effected by the
coordinate system and in some cases may no longer appear as a perfect ellipse.
See the Developer Reference for a code example.

LegacyArc
The LegacyArc object is a portion of an Ellipse and is defined by a DRect, a start angle,
and an end angle. The Ellipse is constructed to be inscribed in the rectangle defined by
the DRect. The rectangle, in which the Ellipse is inscribed, is axis-aligned and is always
rectangular regardless of the coordinate system used. The angles are measured in
degrees with zero being along the positive X-axis and positive angles being in the
counterclockwise direction. The angles are only stored to a tenth of a degree resolution
with values between 0.0 and 360.0.

Pitney Bowes 100 Developer Guide

 6 – Spatial Objects and Coordinate Systems

Because LegacyArc objects are defined by two points (for the DRect) and angles, and are
always displayed axis aligned, they are designed to be used primarily for cosmetic display
purposes. While many operations are available using LegacyArc objects, internally, a
MultiCurve copy of the LegacyArc is used for these operations. This can sometimes lead
to unexpected results.
See the Developer Reference for a code example.

LegacyText
The LegacyText object is the MapInfo Professional equivalent of a text object. If a given
database does not support Text the LegacyText object can be lost when using that format.
LegacyText objects are placed within a geographically-sized rectangle with a lower-left
anchor point specified. The point-size of the text is based upon what fits best within the
rectangle.
LegacyText objects do not fit nicely into the Geometry model. Several methods available
on the Geometry FeatureGeometry classes, such as Combine, make no sense for
LegacyText and will throw a NotSupportedException. Text objects do exist in MapInfo
native TAB files in the Geometry column. The LegacyText class provides a way to access
these objects. Refer to online reference for specific behaviors of LegacyText objects.

Geometry Objects
Geometry objects that are not also FeatureGeometry objects need to be converted to a
suitable FeatureGeometry object to be displayed on a map. Most FeatureGeometry
classes contain constructors that take appropriate Geometry objects and create new
FeatureGeometry objects:
using MapInfo.Geometry;

Curve curve = new Curve(csys, lineString);

MultiCurve multiCurve = new MultiCurve(curve.CoordSys, curve);

The code above creates the Curve using parameters defined elsewhere in the code of a
CoordSys (csys) and a LineString (lineString). A new MultiCurve is then created using
the CoordSys property of the Curve and the Curve itself.
In the example above, as in all FeatureGeometries created from objects, a copy of the
original object is created because the reference cannot be shared.

Pitney Bowes 101 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Curve
The Curve class inherits from the CurveSegmentList class, and represents a contiguous
linear Geometry. Curves contain a collection of CurveSegments that must remain
contiguous. This class is included in the model to allow for future expansion and is part of
the OGC standards.
Use the following example code to model the creation of a Curve:
using MapInfo.Geometry;

DPoint[] points = new DPoint[4];

points[0]= new DPoint(-88.135215,43.998892);

points[1]= new DPoint(-104.875119,43.998892);
points[2]= new DPoint(-120.242895,47.048364);
points[3]= new DPoint(-89.135215 46.998892);

LineString lineString = new LineString(csys, points);

Curve curve = new Curve(csys, lineString);

CurveSegments
At present a CurveSegment can only be a LineString. The class is designed to expand in
future iterations of the product to include Spline, CircularArc, and EllipticalArc
CurveSegments. Curves and Rings are comprised of CurveSegments.

Rings
A Ring is a collection of CurveSegments which must remain contiguous and closed.
Use the following example code to model the creation of a Ring:
using MapInfo.Geometry;

dPoints = new DPoint[102];

dPoints[0] = new DPoint(-109.171279,49.214879);
dPoints[1] = new DPoint(-109.169283,49.241794);
...
dPoints[101] = new DPoint(-109.171279,49.214879);
Ring newRing = new Ring(longLatNad83, CurveSegmentType.Linear, dPoints);

Polygon
A Polygon is an object made up of Rings. A polygon must have at least a single Ring
which defines the exterior boundary of the Polygon. Other Rings can be included inside
which then define holes in the Polygon. Once a Ring is placed inside of another Ring the
object becomes a MultiPolygon.
Use the following example code to model the creation of a Polygon.

Pitney Bowes 102 Developer Guide

 6 – Spatial Objects and Coordinate Systems

using MapInfo.Geometry;

DPoint[][] points = new DPoint[1][];

points[0] = polyPointArrays[0];
Polygon polygon = new Polygon
(longLatNad83, CurveSegmentType.Linear, polyPointArrays[0]);

Checking for Points in Polygons

The following code example shows how to determine whether a point is inside the
boundary of a FeatureGeometry (Multipolygon), on the boundary line or falls outside of it.
Public Shared Sub MapInfoGeometryContainsPoint()
Dim coordSysFactory As CoordSysFactory = Session.Current.CoordSysFactory
Dim coordSys As CoordSys = _
coordSysFactory.CreateLongLat(MapInfo.Geometry.DatumID.NAD83)

Dim points(6) As DPoint

points(0) = New DPoint(-0.705036, -0.122302)
points(1) = New DPoint(-0.446043, 0.486811)
points(2) = New DPoint(0.235012, 0.36211)
points(3) = New DPoint(0.422062, -0.304556)
points(4) = New DPoint(-0.244604, -0.71223)
points(5) = New DPoint(-0.705036, -0.122302)
Dim multiCurve As MultiCurve = New _
MultiCurve(coordSys,CurveSegmentType.Linear,points)
Dim multiPolygon As MultiPolygon = New _
MultiPolygon(coordSys,CurveSegmentType.Linear,points)

Dim insidePoint As DPoint = New DPoint(-0.115108,0.160671)

Dim boundaryPoint As DPoint = New DPoint(-0.446043,0.486811)
Dim outsidePoint As DPoint = New DPoint(-1.103118,0.021583)

If multiPolygon.ContainsPoint(insidePoint) Then _
Console.WriteLine("Points inside area inclosed by closed _
(GeometryDimension 2) objects are contained")
End If
If Not multiCurve.ContainsPoint(insidePoint) Then _
Console.WriteLine("But this is not true for linear _
(GeometryDimension 1) objects")
End If
If multiPolygon.ContainsPoint(boundaryPoint) Then _
Console.WriteLine("Points on the boundary of closed objects _
are contained")
End If
If multiCurve.ContainsPoint(boundaryPoint) Then _
Console.WriteLine("Points lying on linear objects are contained")
End If

Pitney Bowes 103 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

If Not multiPolygon.ContainsPoint(outsidePoint) Then _

Console.WriteLine("Point completely outside closed objects _
are not contained")
End If
If Not multiCurve.ContainsPoint(outsidePoint) Then _
Console.WriteLine("Point completely outside linear objects _
are not contained")
End If
End Sub

Coordinate Systems
Coordinate systems describe the domain in which a particular object or set of objects
reside. The coordinate system allows for the delineation, in specific terms, of the object or
objects being described. The CoordSys classes contain methods, properties and
interfaces that allow for the creation, manipulation, and editing of coordinate systems.
When Geometries are created, they are created in a particular coordinate system
specified in the creation of the object. Objects cannot change the coordinate system in
which they were created. They can only be copied into another coordinate system.
The CoordSys class facilitates the creation and manipulation of coordinate systems. The
Coordsys class uses an XML version of the projection file (C:\Program
Files\MapInfo\Professional).
The CoordSysFactory object contains registered coordinate systems. CoordSys
definitions can be registered by loading one or more XML projection files or by using the
RegisterCoordSys, or RegisterCoordSysInfo methods. Create CoordSys objects from the
factory, or code-codespace (EPSG, SRID), PRJ string, MapBasic string, and other
Factory creation methods. There are also Military Grid Reference System conversion
methods in the CoordSys class.

Creating a CoordSys Object

The following sample code shows the creation of CoordSys objects several different
ways: using a MapInfo codespace; through EPSG; as longitude/latitude from a PRJ
string; from a MapBasic string; and through SRID.
VB example:
Public Shared Sub MapInfoGeometryCreateCoordSys()
Dim factory As CoordSysFactory = Session.Current.CoordSysFactory

’ create CoordSys objects from srsName

Dim csysWGS84 As CoordSys = factory.CreateCoordSys("EPSG:4326")

Pitney Bowes 104 Developer Guide

 6 – Spatial Objects and Coordinate Systems

Dim csysNAD83 As CoordSys = factory.CreateCoordSys_

("mapinfo:coordsys 1,74")
Dim csysNAD27 As CoordSys = factory.CreateCoordSys("SRID:8260")

’ create CoordSys objects from code/codeSpace

csysWGS84 = factory.CreateCoordSys("4326", CodeSpace.Epsg)
csysNAD83 = factory.CreateCoordSys("coordsys 1,74", CodeSpace.MapInfo)
csysNAD27 = factory.CreateCoordSys("8260", CodeSpace.Srid)

’ create CoordSys objects from user-defined parameters

Dim dat As Datum = factory.CreateDatum(DatumID.WGS84)
csysWGS84 = factory.CreateCoordSys(CoordSysType.LongLat, _
dat, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, nothing)
dat = factory.CreateDatum(DatumID.NAD83)
csysNAD83 = factory.CreateCoordSys(CoordSysType.LongLat, _
dat, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, nothing)
dat = factory.CreateDatum(DatumID.NAD27ContinentalUS)
csysNAD27 = factory.CreateCoordSys(CoordSysType.LongLat, _
dat, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, nothing)
’ create Long/Lat coordinate system
csysWGS84 = factory.CreateLongLat(DatumID.WGS84)
csysNAD83 = factory.CreateLongLat(DatumID.NAD83)
csysNAD27 = factory.CreateLongLat(DatumID.NAD27ContinentalUS)

’ create from MapBasic string

Dim csysRGF93 As CoordSys = _
factory.CreateFromMapBasicString("CoordSys Earth Projection 3, _
33, ""m"", 3, 46.5, 44, 49, 700000, 6600000")
’ create from PRJ string
csysNAD83 = factory.CreateFromPrjString("1, 74")
Sub

Changing the Coordinate System of a Geometry Object

The next example illustrates how to convert a Geometry object from one coordinate
system to another.
VB example:
Public Shared Sub MapInfoGeomeTryCoordSys(ByRef coordSys As _
CoordSys, ByRef points() As DPoint, ByRef alternateCoordSys _
as CoordSys
' All Geometry constructors require a CoordSys parameter
' Note that the points array is assumed to be in coordSys
Dim lineString As LineString = New _
LineString(coordSys,points)

' The Geometry has a reference to the CoordSys used during

' construction. Unlike the coordinate data represented by the
' points array, the CoordSys' is not copied
If ReferenceEquals(coordSys, lineString.CoordSys) Then

Pitney Bowes 105 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Console.WriteLine("Geometry objects hold a reference to _

the CoordSys used during construction")
End If

' if you want to convert the object to another coordinate

' system, you need to make a new copy using one of the copy methods
If Not coordSys.Equals(alternateCoordSys) Then
Dim newGeometry as MapInfo.Geometry.Geometry = _
lineString.Copy(alternateCoordSys)
End If
End Sub

Adding Coordinate Systems

If the MapInfoCoordinateSystemSet.xml file does not contain a coordinate system to
match your needs, you may add it, for the library to reference. This feature supports
adding EPSG codes and SRID codes to extend the library’s capabilities.
EPSG codes represent a collection of coordinate systems (known as codespaces)
maintained in the EPSG Geodetic Parameter Dataset under the auspices of the
International Association of Oil & Gas Producers (OPG). The OPG’s Survey and
Positioning Committee took over this responsibility from the European Petroleum Survey
Group in 2005.
SRID codes are unique spatial reference numbers that refer to codespaces for Oracle
Spatial tables.
(MDAL supports a third codespace called MapInfo.)
MDAL provides you with many of the common EPSG and SRID mappings. If you need to
register a different EPSG or SRID code to a particular coordinate system, this feature
provides you with two methods to do so.
To extend MDAL’s ability to use any EPSG or SRID codespace, you may add the
information programmatically, in which case, the coordinate system information will only
last as long as the MDAL Session. Or you may add it to your app.config file as a more
permanent solution. Each is discussed below.

Register EPSG and SRID Codes Programmatically

The MapInfo.Geometry.CoordSysFactory class contains methods that allow you to
register EPSG and SRID codes to a specified coordinate system.
RegisterEPSGCode() and RegisterSRIDCode() each take two parameters: one being the
EPSG or SRID code that represents the codespace, the second is the coordinate system
information that first parameter will map to.

Pitney Bowes 106 Developer Guide

 6 – Spatial Objects and Coordinate Systems

The following example demonstrates registering a fictional code with the Long/Lat NAD83
coordinate system.
VB example:
Public Shared Sub MapInfo_Geometry_RegisterEPSGCode()
Dim factory As CoordSysFactory = Session.Current.CoordSysFactory
' create CoordSys objects from srsName
Dim csysNAD83 As CoordSys = _
factory.CreateCoordSys("mapinfo:coordsys 1,74")
' 9998 is a fictional code for demonstration purposes
Try
factory.RegisterEPSGCode(9998, csysNAD83)
Catch ae As ApplicationException
'code already exists. Codes cannot be duplicated
End Try
End Sub

 If the EPSG or SRID code already exists, an exception will be thrown indicating this
fact.

To determine if a coordinate system for the MapInfo, EPSG or SRID codespace is already
supported, call this method:
• MapInfo.Geometry.CoordSys.Code(codespace).
This method returns the first (or only) occurrence of the codespace that matches or null, if
it does not exist.
Similarly, to return the first SRSName in the list that matches the input codespace, call
this method:
• MapInfo.Geometry.CoordSys.SRSName(codespace).
An SRSName (Spatial Reference System) represents the name of a coordinate reference
system written in GML (Geography Markup Language). This is typically, a friendly name
for the coordinate system, not a list of parameter values.
To get a list of all the codes and coordinate systems that are mapped to a particular
coordinate system, MDAL provides the following methods:
• MapInfo.Geometry.CoordSys.Codes(codeSpace)
• MapInfo.Geometry.CoordSys.SrsNames(codeSpace)

Keep in mind that the coordinate system information you added programmatically, will
only be maintained during the lifetime of the MDAL Session.

Pitney Bowes 107 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

Register EPSG and SRID Codes to a Web or Desktop Configuration File

The second, and more permanent, way to add EPSG or SRID codes to your MDAL app,
is by adding the information to your application app.config1 file. The code below in bold
shows the information to copy and paste into your config file. An explanation of the code
follows.
<?xml version="1.0" encoding="utf-8" ?>
<configuration>
<configSections>
<section name="MapInfo.DataAccess"
type="MapInfo.Engine.ConfigSectionHandler, MapInfo.DataAccess, Version=17.0,
Culture=neutral, PublicKeyToken=93e298a0f6b95eb1" />
</configSections>
<appSettings>
<add key="MapInfo.Engine.Session.UseCallContext" value="false" />
</appSettings>
<MapInfo.CoreEngine>
<EPSG_Code_Mappings>
<EPSG_Code_Mapping>
<srsName>My Custom CRS</srsName>
<srsID>
<code>coordsys 8,74,8,-
110.0833333333,47.5,0.9999375,2624666.667,328083.3333</code>
<codeSpace>mapinfo</codeSpace>
<remarks>My Custom CRS</remarks>
</srsID>
<EPSG_Codes>
<EPSG_Code>9987</EPSG_Code>
<EPSG_Code>9988</EPSG_Code>
<EPSG_Code>9989</EPSG_Code>
</EPSG_Codes>
</EPSG_Code_Mapping>
</EPSG_Code_Mappings>
<SRID_Code_Mappings>
<SRID_Code_Mapping>
<srsName>My Custom CRS</srsName>
<srsID>
<code>coordsys 8,74,8,-
114.0833333333,47.5,0.9999375,2624666.667,328083.3333</code>
<codeSpace>mapinfo</codeSpace>
<remarks>My Custom CRS</remarks>
</srsID>
<SRID_Codes>
<SRID_Code>9990</SRID_Code>
<SRID_Code>9991</SRID_Code>
<SRID_Code>9992</SRID_Code>
</SRID_Codes>

1. If your desktop application does not have an app.config file, you can create one by adding it to your project from
the Visual Studio Application Configuration File template.

Pitney Bowes 108 Developer Guide

 6 – Spatial Objects and Coordinate Systems

</SRID_Code_Mapping>
</SRID_Code_Mappings>
</MapInfo.CoreEngine>
</configuration>

The code above shows that there are two sections of information to add. One is to identify
the correct CoreEngine dll and version number1, the second is to add elements for EPSG
and SRID code mappings.
An EPSG code mapping consists of an SRSName, SRS ID, and EPSG code(s). The
SRSID is further defined by the parameters and codespace for the coordinate system.
An SRID code mapping is similar to the EPSG code mapping, except it refers to the
Oracle Spatial identification number.

1. To determine the correct version number for the MapInfo.DataAccess assembly, examine the properties of the
installed MapInfo.DataAccess.dll (e.g., "C:\Program Files\MapInfo\Professional\MapInfo.DataAccess.dll")

Pitney Bowes 109 Developer Guide

7 – Working with GeoPackage
7
MDAL provides support for opening, displaying, creating and editing
GeoPackage files which is an open source format created by OGC.

In this chapter:
 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111
 Opening a GeoPackage file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111
 Opening a GeoPackage Tab file . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
 Enable GeoPackage as cache for RDB (SQL/Oracle) tables. . . . . 112
 Create and Save GeoPackage file programmatically . . . . . . . . . . . 113
 7 – Working with GeoPackage

Overview
GeoPackage is an open format for Geospatial Information defined by OGC
(http://www.geopackage.org). It is a SQLite-based extension defined by the OGC to
promote portability of data across platforms and products.
According to OGC definition for GeoPackage:
“A GeoPackage is a platform-independent SQLite database file that may contain:
• vector geospatial features
• tile matrix sets of imagery and raster maps at various scales
• metadata
Since a GeoPackage is a database, it supports direct use, meaning that its data can
be accessed and updated in a "native" storage format without intermediate format
translations. GeoPackages are interoperable across all enterprise and personal
computing environments, and are particularly useful on mobile devices like cell phones
and tablets in communications environments with limited connectivity and bandwidth.
This OGC® Encoding Standard defines the schema for a GeoPackage, including table
definitions, integrity assertions, format limitations, and content constraints. The
allowable content of a GeoPackage is entirely defined in this specification."

Opening a GeoPackage file

GeoPackage table and GeoPackage Tab files can be opened programmatically.

 If the GeoPackage table being opened have a Coordinate System which is not
supported by MapInfo Pro then that table will not be opened.

To open a GeoPackage file programmatically use the code below:

TableInfoGeopackage infoGeoPackage = new TableInfoGeopackage(“table alias”);
infoGeoPackage.DatabasePath = tablePath; //Path to the GeoPackage file .gpkg
infoGeoPackage.DatabaseTableName = “Table Name”; //Table to be opened in
GeoPackage
TableInfo tableInfo = infoGeoPackage;//Create the Table Info object.
Table
openTable=MapInfo.Engine.Session.Current.Catalog.OpenTable(tableInfo);//Open
table

A GeoPackage database may contain multiple tables so the database table name is
required.

Pitney Bowes 111 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

 MDAL only supports Feature Tables of GeoPackage.

The following code sample illustrates getting the list of tables in the GeoPackage database:
GeopackageDataSourceDefinition gpkgDSDef = new
GeopackageDataSourceDefinition(“c:\data\test.gpkg”); //path to gpkg file
GeopackageDataSource gpkgDS =
GeopackageDataProvider.Instance.OpenDataSource(gpkgDSDef, null) as
GeopackageDataSource;
List<IDataSourceNamedTable> tablesList = gpkgDS.GetSchemaNamedTables("main",
new string[] { "" }) as List<IDataSourceNamedTable>; //Get the list of tables
in database.

Opening a GeoPackage Tab file

A GeoPackage tab file, which references a Feature Table by name within a GeoPackage
database, may be created using MDAL (programmatically) or MapInfo Pro. Multiple tab
files may reference to the same database since one database may contain multiple
Feature Tables.
To open a GeoPackage Tab file, use the code below:
TableInfo infoGeoPackage = TableInfoGeopackage.CreateFromFile(“Path to Tab
file”);
Table openedTable = Session.Current.Catalog.OpenTable(infoGeoPackage);

Opened table will have table type as GeoPackage

openedTable.TableInfo.TableType == TableType.Geopackage

A GeoPackage tab file created using MapInfo Pro can be opened in MDAL and vice
versa.

Enable GeoPackage as cache for RDB

(SQL/Oracle) tables
You have the option to use GeoPackage as Cache for RDB Tables. Following Metadata
properties should be configured for to enable this functionality.
TableInfoServer tis = new TableInfoServer("GeoPackageCacheAPITest");
tis.ConnectString = "Driver={SQL Server Native Client
11.0};DATABASE=QADB;Server=ServerName;UID=UserName;PWD=Password";//
"DSN=ServerDSN ";
tis.Query = "Select * From Table";

Pitney Bowes 112 Developer Guide

 7 – Working with GeoPackage

tis.Toolkit = ServerToolkit.Odbc;
CacheParameters cp = new CacheParameters(CacheOption.All);
cp.StorageType = CacheStorageType.Geopackage; //Cache Type as Geoackage.
tis.CacheSettings = cp;

Multiple RDB tables are cached in the same GeoPackage Cache database. The benefit
for using the GeoPackage for the cache is that MDAL will use fewer temporary files which
is important for environments where the number of available file handles becomes
limiting.

 Note: When using GeoPackage as a cache format, MDAL will store information that
is not supported by the standard (such as coordinate system information and styles
information). It is advisable not to use Cache file directly.

Create and Save GeoPackage file

programmatically
A GeoPackage table can be created using the TableInfoGeopackage API. This will allow
you to create and save a .gpkg file along with its Tab file at the specified path.
TableInfoGeopackage tig = new TableInfoGeopackage("GeoPackageTest");
tig.Temporary = true;
Column col = new Column();
tig.Columns.Add(ColumnFactory.CreateIndexedIntColumn("ROWNUM"));
tig.Columns.Add(ColumnFactory.CreateStringColumn("StrName", 40));
tig.Columns.Add(ColumnFactory.CreateIndexedDoubleColumn("Doubleval"));
CoordSysFactory csf = new CoordSysFactory();
CoordSys wgs84 = csf.CreateLongLat(DatumID.WGS84);
tig.Columns.Add(ColumnFactory.CreateFeatureGeometryColumn("GEO", wgs84));
tig.Columns.Add(ColumnFactory.CreateStyleColumn());
tig.DatabasePath = _tempPath + "GeoPackageTest.gpkg";
tig.DatabaseTableName = "GeoPackageTest";
tig.TablePath = _tempPath + "GeoPackageTest.TAB";
Table _miTable = Session.Current.Catalog.CreateTable(tig);

The above code will create a GPKG database along with its TAB file and will save that file
at TablePath. If GPKG database already exists then the table will be added to the existing
Database.
Once the table gets created Features can be inserted, updated and deleted from the
table. MDAL can only create GeoPackage tables for which EPSG code and OGC “Well
Known Text” description is available for the Coordinate System. The list of supported
Coordinate Systems in MDAL can be found in “MapInfoCoordinateSystemSet.xml” at
installation path “C:\Program Files\MapInfo\Professional”.

Pitney Bowes 113 Developer Guide

MapInfo Data Access Library 1.0 Developer Guide

 Note: As per GeoPackage Specification Requirements - " Every feature table or

view in a GeoPackage SHALL have a column with column type INTEGER and
PRIMARY KEY AUTOINCREMENT column constraints".

Due to the rule above when a new GeoPackage Table is created, an ID column is added
to the column list by default. So the column order of the newly created table may differ
from the order supplied.
The GeoPackage standard does not support capture of styling information with the data.
As a result, MDAL will apply a default styling to the data. Users should prefer to use layer
style overrides and/or themes on their GeoPackage feature layers to display the data in
the desired way.
MapInfo Pro will store default styling in the .TAB file which MDAL will recognize and apply
it on the GeoPackage table.

Pitney Bowes 114 Developer Guide