ODBC - HEAVY.AI Docs

Open Database Connectivity (ODBC) is a standard application programming interface (API) for accessing database management systems (DBMS). HeavyDB supports ODBC connections.
The HEAVY.AI ODBC installer tool filename has the following format:
HeavyaiInstall_<major_version>.<minor_version>.<minor_patch_version>.<odbc_driver_version>.<odbc_driver_minor_version>.
<odbc_patch_version>.<odbc_extra_info>.zip
For example:
HeavyaiInstall_6.0.0.zip
For assistance with downloading HEAVY.AI ODBC software and utilities, contact your HEAVY.AI Sales Representative.
Connecting from Microsoft Windows Using ODBC
While there are many ways to connect to ODBC, perhaps the most common is to connect from a Windows 10 client using Tableau or Excel. HEAVY.AI provides an ODBC connection utility to help you get started.
To configure your ODBC datasource:
1.Run the HEAVY.AI Installer Tool provided by your HEAVY.AI Sales Representative.
2.Enter the path to the ODBC driver (default is C:\Heavyai\ODBC\bin\).
3.Click OK.
4.Wait for the “SUCCESS!” message. Click OK.
5.Open your Windows control panel.
6.Open either the 32-bit or 64-bit version of ODBC Data Source Administrator, depending on the application you are connecting to the driver.
7.Click Add....
8.Enter the user Name (for example, heavyai), Platform, and Driver (HeavyaiDriver).
9.In HEAVY.AI ODBC Driver DSN Setup:
1.Enter an optional Description for your data source.
2.Enter the User (for example, heavyai).
3.Enter the Password for your data source.
4.Enter the Host:Port values (for example, myhost.heavyai.com:6274).
5.Enter the Database name (for example, heavyai).
6.Enter the Max rows returned (for example, 10000).
7.Select the Protocol Type to connect to the HEAVY.AI Web server: HTTP, HTTPS, Binary, or Binary Encrypted. If you select HTTPS or Binary Encrypted, you must enable encrypted connections using the Microsoft Management Console.
8.If you want to ease server certificate validation requirements, select Ignore Invalid Certificates (not recommended). If you do this, the CN name in the certificate is not required to match the network name of the host from which it was received.
9.In Connection Timeout, enter the time interval (in seconds) used to terminate the failed connection attempts.
10.To configure options for HEAVY.AI integration with applications such as Alteryx, click Advanced Options. You can:
Escape single quotes or line-feed characters when a parameterized insert command is used.
Allow an SQLPrepare command to be run before an SQL SELECT command to return metadata about the columns.  This adds an extra call to the HEAVY.AI database and should be used with caution.
Allow the driver to remove blank schema names from qualified table names. The Alteryx database discovery GUI tool attempts to insert a blank query into qualified names such as heavyai.table. This option allows the driver to remove the extra period, transforming the name to  heavyai.table. 
10.Click Test.
11.Wait for the “Success!” message. Click OK.
Using HTTPS and Binary Encryption
In HEAVY.AI ODBC Driver DSN Setup, if you set your protocol type to HTTPS or Binary Encrypted, follow these instructions to enable an encrypted connection:
1.Open the Microsoft Management Console (MMC).
2.On the File menu, select Add/Remove Snap-ins.
3.In the Available snap-ins list box, select Certificates.
4.In the Certificates snap-in box, select Computer Account, and then click Next.
5.Select Local Computer, and click Finish, and in the Add or Remove Snap-ins window, click OK.
6.In the left pane of the Console Root window, right-click the Trusted Root Certificate Authorities folder, click All tasks, and then click Import....
7.In the Certificate Import Wizard, click Next.
8.Click Browse, select the certificate file to import, and click Finish.
Installing ODBC on Linux
Follow these steps to configure HeavyDB ODBC connections on Linux.
1.If required, install unixODBC, which you use to test that the HEAVY.AI ODBC driver configuration is working correctly:
For RHEL/Centos:
$ sudo yum install unixODBC
For Ubuntu/Debian:
 $ sudo apt-get install unixodbc
Type y when prompted to install the package.
2.Validate that isql is installed by running it from the shell:
$ isql
If installed correctly, you see output showing the isql syntax and options.
3.Create a folder for the ODBC driver:
$ sudo mkdir -p /apps/odbc/heavyai/
$ sudo chmod 777 /apps/odbc/heavyai
4.Change to the folder you just created:
$ cd /apps/odbc/heavyai/
$ pwd /apps/odbc/heavyai
5.Download the HEAVY.AI Linux ODBC driver:
$ curl -O https://builds.mapd.com/odbc/heavydbODBC-6.0.0-20200302-a280de1b9a-Linux.tar.gz -u <username>:<password>
6.
$ tar xvf mapd_odbc_installer_linux_.tar.gz
7.Edit the /etc/odbc.ini file by changing the properties and values to values appropriate for your HEAVY.AI installation. The included /configuration/odbc.ini.skeleton file (show below) provides a template and information about available options.
[ODBC Data Sources]
heavyai=heavyaiDriver
​
[Heavy.AI]
Description=64-bit HEAVY.AI Driver
Driver=/apps/odbc/heavyai/lib/libheavyaiODBC.so
Locale=en-US
PWD=<USER_PASSWORD>
UID=<USER_ID>
# HOST can contain ':port number' as in 'localhost:6274'.
HOST=<SERVER_HOSTNAME>
PORT=<SERVER_PORT>
DATABASE=<DATABASE_NAME>
# Protocol can be BINARY or HTTP.
# If not supplied, the protocol defaults to BINARY.
PROTOCOL=<PROTOCOL>
# If CA_CERT points to a valid PKI certificate, the driver
# attempts to establish an SSL connection. If not supplied
# or empty, the driver defaults to a nonencrypted connection.
# Note that the server port must be expecting an SSL connection.
SERVER_CA_CERT=<PATH_TO_SERVER_CA_CERT>
# Provide a time interval (in seconds) used to terminate the failed connection attempts. 
CONNECTION TIMEOUT=<seconds>
If SERVER_CA_CERT is supplied, it overrides the default search path.
Invalid non-mandatory options--for example, those that have misspellings--are ignored and do not generate errors.
8.Edit the /etc/odbcinst.ini file by changing the following properties/values in bold to values appropriate for your HEAVY.AI installation.
[ODBC] 
Trace = yes 
TraceFile=<FULL_NAME_ODBC_TRACE_FILE> 
​
[ODCB Drivers] 
HeavyaiDriver=Installed 
​
[HeavyaiDriver]
APILevel=1 
DriverODBCVer=04.80 
SQLLevel=1 
ConnectionFunction=YYY 
Description=HeavyaiDriver
Driver=apps/odbc/heavyai/HeavyaiODBC-6.0.0.prod-4.0.000.0000-20200302-a280de1b9a-Linux/heavyai/odbc/lib/libheavyaiODBC.so 
LogLevel=6 
LogPath=<PATH_TO_HEAVYAI_LOGFILE>
Your odbcinst.ini file might be empty or might already contain other entries. If your file contains other entries, add the new entries to the end of the file and do not overwrite existing entries.
9.Create a symbolic link to the error messages folder for the HEAVY.AI driver library:
$ ln -s apps/odbc/heavyai/HeavyaiODBC-6.0.0.prod-4.0.000.0000-20200302-a280de1b9a-Linux/heavyai/odbc/lib/libheavyaiODBC.so
$ ls -l /apps/odbc/heavyai/libs/
total 229256
lrwxrwxrwx 1 mapd mapd        60 Feb  5 12:54 en-US -> 
/apps/odbc/heavyai/DataAccessComponents/ErrorMessages/en-US/
-rwxrwxr-x 1 mapd mapd 234750688 Oct 11 15:18 libheavyaiODBC.so
10.Test the installation and connection using isql:
$ isql -v heavyai
+---------------------------------------+
| Connected!                            |
|                                       |
| sql-statement                         |
| help [tablename]                      |
| quit                                  |
|                                       |
+---------------------------------------+
SQL> select count(*) cnt from flights_2008_7M;
+---------------------+
| BIGINT              |
+---------------------+
| 7009728             |
+---------------------+
SQLRowCount returns 1
1 rows fetched
SQL
If you receive an error message, your connection is not successfully configured. Check the connection properties in /etc/odbc.ini.
If SERVER_CA_CERT is supplied, it overrides the default search path. Invalid nonmandatory options--for example, those that have misspellings--are ignored and do not generate errors.
Supported ODBC Functions
HeavyDB supports the following Tier 1 (Core) ODBC functions:
Function
Description
​SQLAllocHandle​
Obtains an environment, connection, statement, or descriptor handle.
​SQLBindCol​
SQLBindCol binds application data buffers to columns in the result set.
​SQLCloseCursor​
SQLCloseCursor closes a cursor that has been opened on a statement and discards pending results.
​SQLColAttribute​
SQLColAttribute returns descriptor information for a column in a result set. Descriptor information is returned as a character string, descriptor-dependent value, or an integer value.
​SQLColumns​
SQLColumns returns the list of column names in specified tables. The driver returns this information as a result set on the specified StatementHandle.
​SQLConnect​
SQLConnect establishes a connection between a driver and a data source. The connection handle references storage of all information about the connection to the data source, including status, transaction state, and error information.
​SQLCopyDesc​
SQLCopyDesc copies descriptor information from one descriptor handle to another.
​SQLDescribeCol​
SQLDescribeCol returns the result descriptor for one column in the result set.
​SQLDescribeParam​
SQLDescribeParam returns the description of a parameter marker associated with a prepared SQL statement. This information is also available in the fields of the IPD.
​SQLDisconnect​
SQLDisconnect closes the connection associated with a specific connection handle.
​SQLDriverConnect​
SQLDriverConnect is an alternative to SQLConnect.It supports data sources that require more connection information than the three arguments in SQLConnect, dialog boxes to prompt the user for all connection information, and data sources that are not defined in the system information.
​SQLExecDirect​
SQLExecDirect executes a preparable statement using the current values of the parameter marker variables, if any parameters exist in the statement. SQLExecDirect is the fastest way to submit an SQL statement for one-time execution.
​SQLExecute​
SQLExecute executes a prepared statement using the current values of the parameter marker variables, if any parameter markers exist in the statement.
​SQLFetch​
SQLFetch fetches the next rowset of data from the result set and returns data for all bound columns.
​SQLFreeHandle​
SQLFreeHandle frees resources associated with a specific environment, connection, statement, or descriptor handle.
​SQLFreeStmt​
SQLFreeStmt stops processing associated with a specific statement, closes any open cursors associated with the statement, discards pending results, or, optionally, frees all resources associated with the statement handle.
​SQLGetConnectAttr​
SQLGetConnectAttr returns the current setting of a connection attribute.
​SQLGetData​
SQLGetData retrieves data for a single column in the result set, or for a single parameter after SQLParamData returns SQL_PARAM_DATA_AVAILABLE. You can call it multiple times to retrieve variable-length data in parts.
​SQLGetDescField​
SQLGetDescField returns the current setting or value of a single field of a descriptor record.
Unsupported ODBC Functions
HeavyDB does not support the following ODBC functions at this time.
SQLBindParameter
SQLBrowseConnect
SQLCancel
SQLCancelHandle
SQLColumnPrivileges
SQLEndTran
SQLExtendedFetch
SQLFetchScroll
SQLForeignKeys
SQLGetCursorName
SQLSetCursorName
SQLSetPos
SQLSpecialColumns
SQLParamData
SQLPrepare
SQLPrimaryKeys
SQLProcedureColumns
SQLProcedures
SQLPutData
SQLTablePrivileges
Unsupported ODBC Features
HeavyDB does not support the following ODBC features at this time.
Batch statements
Multiple result sets
Domains
Rules
Database procedures
Indexes
Keys
Transactions
Schemas (any client can see tables and views created by any user without restriction)
Rollbacks, checkpoints, or any other type of database recovery
Installing the Power BI Data Connector
To use the HEAVY.AI analytics engine, the Power BI Desktop connects via the HEAVY.AI ODBC driver in DirectQuery mode. (Power BI has two modes: direct and import.)
The Power BI Data Connector is distributed with the ODBC driver; install it to the directory that the Power BI desktop instance defines for custom connectors. To connect in DirectQuery mode, the ODBC driver connection needs to be "wrapped" in the supplied Power BI Data Connector.
Install the Connector
To install the connector, copy HeavyaiPBIC.pqx to the Power BI Desktop custom connectors folder. For a standard Power BI install, this directory is \Users\<user name>\Documents\Power BI Desktop\Custom Connectors. You might need to create the directory.
Security
HeavyaiPBIC.pqx is a signed file. For the signature to be trusted, the thumbprint of the signing certificate needs to be added to the "TrustedCertificateThumbprints” registry entry in the HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Power BI Desktop path. The thumbprint of the signing key is 29A21397B8169B153577806E3C43A502D944BC99.
For more information, see Trusted third-party connectors.
Alternatively, you can downgrade security on Power BI desktop to allow it to trust all connectors.

APIs and Interfaces - Previous

JDBC

Next - APIs and Interfaces

Vega

Last modified 4mo ago

Export as PDF

Copy link

Outline

Connecting from Microsoft Windows Using ODBC

Using HTTPS and Binary Encryption

Installing ODBC on Linux

Supported ODBC Functions

Unsupported ODBC Functions

Unsupported ODBC Features

Installing the Power BI Data Connector

Install the Connector

Security