An asynchronous/synchronous interface for node.js to unixODBC and its supported drivers.
- unixODBC binaries and development libraries for module compilation
- on Ubuntu/Debian
sudo apt-get install unixodbc unixodbc-dev
- on RedHat/CentOS
sudo yum install unixODBC unixODBC-devel
- on OSX
- using macports.org
sudo port unixODBC
- using brew
brew install unixODBC
- using macports.org
- on IBM i
yum install unixODBC unixODBC-devel
(requires yum)
- on Ubuntu/Debian
- odbc drivers for target database
- properly configured odbc.ini and odbcinst.ini.
After insuring that all requirements are installed you may install by one of the two following options:
git clone git://github.com/wankdanker/node-odbc.git
cd node-odbc
node-gyp configure build
npm install odbc
var db = require('odbc')()
, cn = process.env.ODBC_CONNECTION_STRING
;
db.open(cn, function (err) {
if (err) return console.log(err);
db.query('select * from user where user_id = ?', [42], function (err, data) {
if (err) console.log(err);
console.log(data);
db.close(function () {
console.log('done');
});
});
});
The simple api is based on instances of the Database
class. You may get an
instance in one of the following ways:
require("odbc").open(connectionString, function (err, db){
//db is already open now if err is falsy
});
or by using the helper function:
var db = require("odbc")();
or by creating an instance with the constructor function:
var Database = require("odbc").Database
, db = new Database();
Returns a Boolean of whether the database is currently connected.
var db = require("odbc")();
console.log( "Connected: " + db.connected );
Open a connection to a database.
- connectionString - The ODBC connection string for your database
- callback -
callback (err)
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
db.open(cn, function (err) {
if (err) {
return console.log(err);
}
//we now have an open connection to the database
});
Synchronously open a connection to a database.
- connectionString - The ODBC connection string for your database
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
try {
var result = db.openSync(cn);
}
catch (e) {
console.log(e.message);
}
//we now have an open connection to the database
Issue an asynchronous SQL query to the database which is currently open.
- sqlQuery - The SQL query to be executed.
- bindingParameters - OPTIONAL - An array of values that will be bound to
any '?' characters in
sqlQuery
. - callback -
callback (err, rows, moreResultSets)
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
db.open(cn, function (err) {
if (err) {
return console.log(err);
}
//we now have an open connection to the database
//so lets get some data
db.query("select top 10 * from customers", function (err, rows, moreResultSets) {
if (err) {
return console.log(err);
}
console.log(rows);
//if moreResultSets is truthy, then this callback function will be called
//again with the next set of rows.
});
});
Synchronously issue a SQL query to the database that is currently open.
- sqlQuery - The SQL query to be executed.
- bindingParameters - OPTIONAL - An array of values that will be bound to
any '?' characters in
sqlQuery
.
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//blocks until the connection is opened.
db.openSync(cn);
//blocks until the query is completed and all data has been acquired
var rows = db.querySync("select top 10 * from customers");
console.log(rows);
Close the currently opened database.
- callback -
callback (err)
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
db.open(cn, function (err) {
if (err) {
return console.log(err);
}
//we now have an open connection to the database
db.close(function (err) {
console.log("the database connection is now closed");
});
});
Synchronously close the currently opened database.
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
//Blocks until the connection is closed
db.closeSync();
Prepare a statement for execution.
- sql - SQL string to prepare
- callback -
callback (err, stmt)
Returns a Statement
object via the callback
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
db.prepare("insert into hits (col1, col2) VALUES (?, ?)", function (err, stmt) {
if (err) {
//could not prepare for some reason
console.log(err);
return db.closeSync();
}
//Bind and Execute the statment asynchronously
stmt.execute(['something', 42], function (err, result) {
result.closeSync();
//Close the connection
db.closeSync();
});
})
Synchronously prepare a statement for execution.
- sql - SQL string to prepare
Returns a Statement
object
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
//Blocks while preparing the statement
var stmt = db.prepareSync("insert into hits (col1, col2) VALUES (?, ?)")
//Bind and Execute the statment asynchronously
stmt.execute(['something', 42], function (err, result) {
result.closeSync();
//Close the connection
db.closeSync();
});
Begin a transaction
- callback -
callback (err)
Synchronously begin a transaction
Commit a transaction
- callback -
callback (err)
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
db.beginTransaction(function (err) {
if (err) {
//could not begin a transaction for some reason.
console.log(err);
return db.closeSync();
}
var result = db.querySync("insert into customer (customerCode) values ('stevedave')");
db.commitTransaction(function (err) {
if (err) {
//error during commit
console.log(err);
return db.closeSync();
}
console.log(db.querySync("select * from customer where customerCode = 'stevedave'"));
//Close the connection
db.closeSync();
});
})
Synchronously commit a transaction
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
db.beginTransactionSync();
var result = db.querySync("insert into customer (customerCode) values ('stevedave')");
db.commitTransactionSync();
console.log(db.querySync("select * from customer where customerCode = 'stevedave'"));
//Close the connection
db.closeSync();
Rollback a transaction
- callback -
callback (err)
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
db.beginTransaction(function (err) {
if (err) {
//could not begin a transaction for some reason.
console.log(err);
return db.closeSync();
}
var result = db.querySync("insert into customer (customerCode) values ('stevedave')");
db.rollbackTransaction(function (err) {
if (err) {
//error during rollback
console.log(err);
return db.closeSync();
}
console.log(db.querySync("select * from customer where customerCode = 'stevedave'"));
//Close the connection
db.closeSync();
});
})
Synchronously rollback a transaction
var db = require("odbc")()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
//Blocks until the connection is open
db.openSync(cn);
db.beginTransactionSync();
var result = db.querySync("insert into customer (customerCode) values ('stevedave')");
db.rollbackTransactionSync();
console.log(db.querySync("select * from customer where customerCode = 'stevedave'"));
//Close the connection
db.closeSync();
The node-odbc Pool
is a rudimentary connection pool which will attempt to have
database connections ready and waiting for you when you call the open
method.
If you use a Pool
instance, any connection that you close will cause another
connection to be opened for that same connection string. That connection will
be used the next time you call Pool.open()
for the same connection string.
This should probably be changed.
Get a Databaseinstance which is already connected to
connectionString`
- connectionString - The ODBC connection string for your database
- callback -
callback (err, db)
var Pool = require("odbc").Pool
, pool = new Pool()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
pool.open(cn, function (err, db) {
if (err) {
return console.log(err);
}
//db is now an open database connection and can be used like normal
//if we run some queries with db.query(...) and then call db.close();
//a connection to `cn` will be re-opened silently behind the scense
//and will be ready the next time we do `pool.open(cn)`
});
Close all connections in the Pool
instance
- callback -
callback (err)
var Pool = require("odbc").Pool
, pool = new Pool()
, cn = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname"
;
pool.open(cn, function (err, db) {
if (err) {
return console.log(err);
}
//db is now an open database connection and can be used like normal
//but all we will do now is close the whole pool
pool.close(function () {
console.log("all connections in the pool are closed");
});
});
var odbc = require("odbc")
, util = require('util')
, db = new odbc.Database()
;
var connectionString = "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname";
db.open(connectionString, function(err) {
db.query("select * from table", function(err, rows, moreResultSets) {
console.log(util.inspect(rows, null, 10));
db.close(function() {
console.log("Database connection closed");
});
});
});
Tests can be run by executing npm test
from within the root of the node-odbc
directory. You can also run the tests by executing node run-tests.js
from
within the /test
directory.
By default, the tests are setup to run against a sqlite3 database which is
created at test time. This will require proper installation of the sqlite odbc
driver. On Ubuntu: sudo apt-get install libsqliteodbc
If you would like to enable debugging messages to be displayed you can add the
flag DEBUG
to the defines section of the binding.gyp
file and then execute
node-gyp rebuild
.
<snip>
'defines' : [
"DEBUG"
],
<snip>
You may also enable the ability to load a specific ODBC driver and bypass the
ODBC driver management layer. A performance increase of ~5Kqps was seen using
this method with the libsqlite3odbc driver. To do this, specify the dynodbc
flag in the defines section of the binding.gyp
file. You will also need to
remove any library references in binding.gyp
. Then execute node-gyp rebuild
.
<snip>
'defines' : [
"dynodbc"
],
'conditions' : [
[ 'OS == "linux"', {
'libraries' : [
//remove this: '-lodbc'
],
<snip>
By default, UNICODE suppport is enabled. This should provide the most accurate way to get Unicode strings submitted to your database. For best results, you may want to put your Unicode string into bound parameters.
However, if you experience issues or you think that submitting UTF8 strings will
work better or faster, you can remove the UNICODE
define in binding.gyp
<snip>
'defines' : [
"UNICODE"
],
<snip>
When converting a database time to a C time one may use timegm
or timelocal
. See
man timegm
for the details of these two functions. By default the node-odbc bindings
use timelocal
. If you would prefer for it to use timegm
then specify the TIMEGM
define in binding.gyp
<snip>
'defines' : [
"TIMEGM"
],
<snip>
When column names are retrieved from ODBC, you can request by SQL_DESC_NAME or SQL_DESC_LABEL. SQL_DESC_NAME is the exact column name or none if there is none defined. SQL_DESC_LABEL is the heading or column name or calculation. SQL_DESC_LABEL is used by default and seems to work well in most cases.
If you want to use the exact column name via SQL_DESC_NAME, enable the STRICT_COLUMN_NAMES
define in binding.gyp
<snip>
'defines' : [
"STRICT_COLUMN_NAMES"
],
<snip>
Be aware that through node v0.9 the uv_queue_work function, which is used to execute the ODBC functions on a separate thread, uses libeio for its thread pool. This thread pool by default is limited to 4 threads.
This means that if you have long running queries spread across multiple instances of odbc.Database() or using odbc.Pool(), you will only be able to have 4 concurrent queries.
You can increase the thread pool size by using @developmentseed's [node-eio] (https://github.com/developmentseed/node-eio).
npm install eio
var eio = require('eio');
eio.setMinParallel(threadCount);
- If you have column names longer than 30 characters, you should add
"TDS_Version=7.0" to your connection string to retrive the full column name.
- Example : "DRIVER={FreeTDS};SERVER=host;UID=user;PWD=password;DATABASE=dbname;TDS_Version=7.0"
- If you got error "[unixODBC][FreeTDS][SQL Server]Unable to connect to data source"
Try use SERVERNAME instead of SERVER
- Example : "DRIVER={FreeTDS};SERVERNAME=host;UID=user;PWD=password;DATABASE=dbname"
- Be sure that your odbcinst.ini has the proper threading configuration for your FreeTDS driver. If you choose the incorrect threading model it may cause the thread pool to be blocked by long running queries. This is what @wankdanker currently uses on Ubuntu 12.04:
[FreeTDS]
Description = TDS driver (Sybase/MS SQL)
Driver = libtdsodbc.so
Setup = libtdsS.so
CPTimeout = 120
CPReuse =
Threading = 0
- Dan VerWeire ([email protected])
- Lee Smith ([email protected])
- Bruno Bigras
- Christian Ensel
- Yorick
- Joachim Kainz
- Oleg Efimov
- paulhendrix
Copyright (c) 2013 Dan VerWeire [email protected]
Copyright (c) 2010 Lee Smith [email protected]
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Version 2.0 is an attempt to update the node-odbc package to improve performance and make it more stable between node versions. Every attempt has been made to maintain the interface so no programs should break.
The following is a list of changes that were made in 2.0. The two major changes are:
- Switched from using NAN to N-API (or more specifically, node-addon-api). N-API is a new ABI-stable library for writing node addons that is officially supported by the node.js group. It doesn't need to be recompiled when using a new node version. Also, async work is abstracted away with the N-API AsyncWorker abstract class (still retain uv library to use the mutex).
- No longer gets data using ODBC method SQLGetData. Instead, column data is bound to buffer arrays and then retrieved using SQLFetch method. This is MUCH faster if getting more than a few columns. Our performance tests showed that 100k columns took 19 seconds using SQLBindCol (new way), and 19 minutes using SQLGetData (old way).
Here are some issues that are unresolved:
-
Not sure that I have correctly accounted for all of the differenct data types when binding parameters or when binding column data to insert or retrieve. Most of these are in odbc.cpp and need a look over.
-
There are a few areas where I wasn't clear on how to convert NAN code to N-API. I have marked these areas with comments that being with 'HELP'.
-
This package has been tested on IBMi with os400 (AIX-like) and a DB2 database, and on a RedHat machine with sqlite. Definitely need a lot more testing before switched to be the main package.
If there are any questions about programming desicions, please create an issue here on github or reach out to me at [email protected]