A Dart package for interacting with ODBC databases. It allows you to connect to ODBC data sources and execute SQL queries directly from your Dart applications.
This package is inspired by the obsolete odbc package by Juan Mellado.
- Instanciate the ODBC class by providing the path to the odbc driver on the host machine
final odbc = DartOdbc(
'/path/to/the/odbc/driver',
version=SQL_OV_ODBC3_80 // optional
);
Path to the ODBC driver can be found in the ODBC driver manager.
In windows this is a .dll
file that is there in the installation folder of the ODBC driver.
In linux this has an extension of .so
.
In macos this should have an extension of .dylib
.
The ODBC version can be specified using the version
parameter.
Definitions for these values can be found in the LibODBC
class.
Please note that some drivers may not work properly with manually setting version.
- Connect to the database by providing the DSN (Data Source Name) configured in the ODBC Driver Manager
await odbc.connect(
dsn: '<your_dsn>',
username: 'db_username',
password: 'db_password',
);
- Or connect to the database via connection string
await odbc.connectWithConnectionString(
"DRIVER={Microsoft Excel Driver (*.xls, *.xlsx, *.xlsm, *.xlsb)};DBQ=C:\Users\Computer\AppData\Local\Temp\test.xlsx;"
);
This is the name you gave when setting up the driver manager. For more information, visit this page from the MySQL Documentation
- In case the path privided to the driver is invalid or there is any issue with setting up the environment/connecting to the database, an
Exception
will be thrown when intanciating the ODBC or connecting to the database. - Execute your queries directly as follows
final result = await odbc.execute("SELECT 10");
- Prepared statements can be used to prevent
SQL Injection
- Example query
final List<Map<String, dynamic>> result = await odbc.execute(
'SELECT * FROM USERS WHERE UID = ?',
params: [1],
);
- The abstraction layer of DartOdbc should be able to handle output for most queries
- But output for columns with very long column size or uncommon data types could get corrupted due to issues in default memory allocation
- Thes can be handled by providing the
ColumnType
in thecolumnConfig
parameter of theexecute
method onDartOdbc
class - Please refer the following example
// Assume a table like this
// +-----+-------+-------------+
// | UID | NAME | DESCRIPTION |
// +-----+-------+-------------+
// | 1 | Alice | |
// | 2 | Bob | |
// +-----+-------+-------------+
// The name is a column of size 150
// The description is a column of size 500
result = await odbc.execute(
'SELECT * FROM USERS WHERE UID = ?',
params: [1],
/// The column config can be provided as this.
/// But for most cases this config is not necessary
/// This is only needed when the data fetching is not working as expected
/// Only the columns with issues need to be provided
columnConfig: {
'NAME': ColumnType(size: 150),
'DESCRIPTION': ColumnType(type: SQL_C_WCHAR, size: 500),
},
);
- Result will be a
Future
ofList
ofMap
objects (Future<List<Map<String, dynamic>>>
) where each Map represents a row. If anything goes wrong anODBCException
will be thrown
final List<Map<String, String>> tables = await odbc.getTables();
- Finally, don't forget to
disconnect
from the database and free resources.
await odbc.disconnect();
-
Native
ODBC
methods can be executed by using theLibOdbc
class -
For more information on the
ODBC
api go to Microsoft ODBC Documentation
This package has been tested to be working on the following Database Servers
- Microsoft SQL Sever
- Oracle
- Although not tested, this plugin should work on any database that provides an
ODBC Driver
. - For a comprehensive list of supported database servers checkout
Drivers
section of the official unixodbc site