18 KiB
SQL Batch API
The SQL Batch API enables you to request queries with long-running CPU processing times. Typically, these kind of requests raise timeout errors when using the SQL API. In order to avoid timeouts, you can use the SQL Batch API to create, read, list, update and cancel queries. You can also run multiple SQL queries in one job. The SQL Batch API schedules the incoming jobs and allows you to request the job status for each query.
The Batch API is not intended to be used for large query payloads than contain over 4096 characters. If you are inserting a large number of rows (over 4kb) into your table, you still need to use the Import API or SQL API for this type of data management. The Batch API is specific to queries and CPU usage.
Note: In order to use the SQL Batch API, your table must be public, or you must be authenticated using API keys. Review the Usage Notes for best practices about using the SQL Batch API.
SQL Batch API Job Schema
The SQL Batch API request to your CartoDB account includes the following job schema elements. Only the query
element can be modified. All other elements of the job schema are defined by the SQL Batch API and are read-only.
Name | Description |
---|---|
job_id |
a universally unique identifier (uuid). |
user |
user identifier, as displayed by the username. |
status |
displays the result of the long-running query. The possible status results are: |
--- | --- |
|_ pending |
job waiting to be executed. |
|_ running |
indicates that the job is currently running. |
|_ done |
job executed successfully. |
|_ failed |
job executed but failed, with errors. |
|_ canceled |
job canceled by user request. |
|_ unknown |
appears when it is not possible to determine what exactly happened with the job. |
query |
the SQL statement to be executed in a database. You can modify the select SQL statement to be used in the job schema. Tip: In some scenarios, you may need to retrieve the query results from a finished job. See Retrieving Job Results for details. |
created_at |
the date and time when the job schema was created. |
updated_at |
the date and time of when the job schema was last updated, or modified. |
failed_reason |
displays the database error message, if something went wrong. |
Example
HEADERS: 201 CREATED; application/json
BODY: {
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: UPDATE airports SET type = 'international'”,
“status”: “pending”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-15T07:36:25Z”
}
Create a Job
To create an SQL Batch API job, make a POST request with the following parameters.
Creates an SQL Batch API job request.
HEADERS: POST /api/v2/sql/job
BODY: {
query: ‘UPDATE airports SET type = 'international'’
}
Response
HEADERS: 201 CREATED; application/json
BODY: {
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: “UPDATE airports SET type = 'international'”,
“status”: “pending”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-15T07:36:25Z”
}
POST Examples
If you are using the Batch API create operation for cURL POST request, use the following code:
curl -X POST -H "Content-Type: application/json" -d '{
"query": "CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)"
}' "http://{username}.cartodb.com/api/v2/sql/job"
If you are using the Batch API create operation for a Node.js client POST request, use the following code:
var request = require("request");
var options = {
method: 'POST',
url: 'http://{username}.cartodb.com/api/v2/sql/job',
headers: { 'content-type': 'application/json' },
body: {
query: 'CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)'
},
json: true
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
Read a Job
To read an SQL Batch API job, make a GET request with the following parameters.
HEADERS: GET /api/v2/sql/job/de305d54-75b4-431b-adb2-eb6b9e546014
BODY: {}
Response
HEADERS: 200 OK; application/json
BODY: {
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: “UPDATE airports SET type = 'international'”,
“status”: “pending”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-15T07:36:25Z”
}
GET Examples
If you are using the Batch API read operation for cURL GET request, use the following code:
curl -X GET "http://{username}.cartodb.com/api/v2/sql/job/{job_id}"
If you are using the Batch API read operation for a Node.js client GET request, use the following code:
var request = require("request");
var options = {
method: 'GET',
url: 'http://{username}.cartodb.com/api/v2/sql/job/{job_id}'
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
List Jobs
To list SQL Batch API jobs, make a GET request with the following parameters.
HEADERS: GET /api/v2/sql/job
BODY: {}
Response
HEADERS: 200 OK; application/json
BODY: [{
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: “UPDATE airports SET type = 'international'”,
“status”: “pending”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-15T07:36:25Z”
}, {
“job_id”: “ba25ed54-75b4-431b-af27-eb6b9e5428ff”,
“user”: “cartofante”
“query”: “DELETE FROM user_dataset”,
“status”: “pending”,
“created_at”: “2015-12-15T07:43:12Z”,
“updated_at”: “2015-12-15T07:43:12Z”
}]
GET Examples
If you are using the Batch API list operation for cURL GET request, use the following code:
curl -X GET "http://{username}.cartodb.com/api/v2/sql/job"
If you are using the Batch API list operation for a Node.js client GET request, use the following code:
var request = require("request");
var options = {
method: 'GET',
url: 'http://{username}.cartodb.com/api/v2/sql/job'
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
Update a Job
To update an SQL Batch API job, make a PUT request with the following parameters.
HEADERS: PUT /api/v2/sql/job/de305d54-75b4-431b-adb2-eb6b9e546014
BODY: {
"query": “SELECT cartodb_id FROM user_dataset”
}
Response
HEADERS: 200 OK; application/json
BODY: {
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: “SELECT cartodb_id FROM user_dataset”,
“status”: “pending”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-17T15:45:56Z”
}
Note: Jobs can only be updated while the status: "pending"
, otherwise the SQL Batch API Update operation is not allowed. You will receive an error if the job status is anything but "pending".
errors: [
“The job status is not pending, it cannot be updated”
]
PUT Examples
If you are using the Batch API update operation for cURL PUT request, use the following code:
curl -X PUT -H "Content-Type: application/json" -d '{
"query": "CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)"
}' "http://{username}.cartodb.com/api/v2/sql/job/{job_id}"
If you are using the Batch API update operation for a Node.js client PUT request, use the following code:
var request = require("request");
var options = {
method: 'PUT',
url: 'http://{username}.cartodb.com/api/v2/sql/job/{job_id}',
headers: {
'content-type': 'application/json'
},
body: { query: 'select the_geom from my_public_dataset' },
json: true
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
Cancel a Job
To cancel an SQL Batch API job, make a DELETE request with the following parameters.
HEADERS: DELETE /api/v2/sql/job/de305d54-75b4-431b-adb2-eb6b9e546014
BODY: {}
Note: Be mindful when cancelling a job when the status: pending
or running
.
- If the job is
pending
, the job will never be executed - If the job is
running
, the job will be terminated immediately
Response
HEADERS: 200 OK; application/json
BODY: {
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: “UPDATE airports SET type = 'international'”,
“status”: “cancelled”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-17T06:22:42Z”
}
Note: Jobs can only be cancelled while the status: "running"
or status: "pending"
, otherwise the SQL Batch API Cancel operation is not allowed. You will receive an error if the job status is anything but "running" or "pending".
errors: [
“The job status is done, cancel is not allowed”
]
DELETE Examples
If you are using the Batch API cancel operation for cURL DELETE request, use the following code:
curl -X DELETE "http://{username}.cartodb.com/api/v2/sql/job/{job_id}"
If you are using the Batch API cancel operation for a Node.js client DELETE request, use the following code:
var request = require("request");
var options = {
method: 'DELETE',
url: 'http://{username}.cartodb.com/api/v2/sql/job/{job_id}',
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
Multi Query Batch Jobs
In some cases, you may need to run multiple SQL queries in one job. The Multi Query batch option enables you run an array of SQL statements, and define the order in which the queries are executed. You can use any of the operations (create, read, list, update, cancel) for the queries in a Multi Query batch job.
HEADERS: POST /api/v2/sql/job
BODY: {
query: [
‘UPDATE airports SET type = 'international'_0’,
‘SELECT * FROM user_dataset_1’,
‘SELECT * FROM user_dataset_2’
]
}
Response
HEADERS: 201 CREATED; application/json
BODY: {
“job_id”: “de305d54-75b4-431b-adb2-eb6b9e546014”,
“user”: “cartofante”
“query”: [{
“query”: “UPDATE airports SET type = 'international'_0”,
“status”: “pending”
}, {
“query”: “SELECT * FROM user_dataset_1”,
“status”: “pending”
}, {
“query”: “SELECT * FROM user_dataset_2”,
“status”: “pending”
}],
“status”: “pending”,
“created_at”: “2015-12-15T07:36:25Z”,
“updated_at”: “2015-12-15T07:36:25Z”
}
Note: The SQL Batch API returns a job status for both the parent Multi Query request, and for each child query within the request. The order in which each query is executed is guaranteed. Here are the possible status results for Multi Query batch jobs:
-
If one query within the Multi Query batch fails, the
"status": "failed"
is returned for both the job and the query, and any "pending" queries will not be processed -
If you cancel the Multi Query batch job, the job status changes to
"status": "cancelled"
. Any running queries within the job will be stopped and changed to"status": "pending"
, and will not be processed -
Suppose the first query job status is
"status": "done"
, the second query is"status": "running"
, and the third query"status": "pending"
. If the second query fails for some reason, the job status changes to"status": "failed"
and the last query will not be processed. It is indicated which query failed in the Multi Query batch job
POST Examples
If you are using the Batch API Multi Query operation for cURL POST request, use the following code:
curl -X POST -H "Content-Type: application/json" -d '{
"query": [
"CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)",
"DROP TABLE airports",
"ALTER TABLE world_airports RENAME TO airport;"
]
}' "http://{username}.cartodb.com/api/v2/sql/job"
If you are using the Batch API Multi Query operation for a Node.js client POST request, use the following code:
var request = require("request");
var options = {
method: 'POST',
url: 'http://{username}.cartodb.com/api/v2/sql/job',
headers: { 'content-type': 'application/json' },
body: {
"query": [
"CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)",
"DROP TABLE airports",
"ALTER TABLE world_airports RENAME TO airport;"
]
},
json: true
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
PUT Examples
If you are using the Batch API Multi Query operation for cURL PUT request, use the following code:
curl -X PUT -H "Content-Type: application/json" -d '{
"query": [
"CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)",
"DROP TABLE airports",
"ALTER TABLE world_airports RENAME TO airport;",
"UPDATE airports SET airport = upper(airport)"
]
}' "http://{username}.cartodb.com/api/v2/sql/job/{job_id}"
If you are using the Batch API Multi Query operation for a Node.js client PUT request, use the following code:
var request = require("request");
var options = {
method: 'PUT',
url: 'http://{username}.cartodb.com/api/v2/sql/job/{job_id}',
headers: { 'content-type': 'application/json' },
body: {
query: [
"CREATE TABLE world_airports AS SELECT a.cartodb_id, a.the_geom, a.the_geom_webmercator, a.name airport, b.name country FROM world_borders b JOIN airports a ON ST_Contains(b.the_geom, a.the_geom)",
"DROP TABLE airports",
"ALTER TABLE world_airports RENAME TO airport;",
"UPDATE airports SET airport = upper(airport)"
]
},
json: true
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});
Retrieving Job Results
In some scenarios, you may need to collect the job results. If that is the case, wrap the query with SELECT * INTO, or CREATE TABLE AS. The results are stored in a new table in your database. For example:
-
A job query,
SELECT * FROM user_dataset;
-
Wrap the query,
SELECT * INTO job_result FROM (SELECT * FROM user_dataset) AS job;
-
Once the table is created, retrieve the results through the CartoDB SQL API,
SELECT * FROM job_result;
Alternatively, use UPDATE
as there is no need to wrap the query. For example, UPDATE airports SET type = 'international'
.
Usage Notes
For best practices, ensure that you are following these recommended usage notes when using the SQL Batch API:
-
The Batch API is not intended for large INSERT jobs, use the Import API for this type of data management
-
Only the
query
element of the job scheme can be modified. All other elements of the job schema are defined by the SQL Batch API and are read-only -
If you are creating a new table from a query without wrapping the query (via INTO or CREATE TABLE AS), the result will create ghost tables that are not configured for CartoDB. Ensure you wrap the query correctly, or use UPDATE instead
-
There is a limit of 4kb per job, or 4096 characters per query. The following limit error message appears if your job exceeds this size:
Your payload is too large (4097). Max size allowed is 4096 (4kb)
- Your table must be public, or you must be authenticated using API keys. The following error message appears if you are using private tables, or are not authenticated:
{"error":["permission denied"]}
For all access to private tables, and for write access to public tables, CartoDB enforces secure API access that requires you to authorize your queries. In order to authorize queries, you can use an API Key or a Consumer Key, as shown in the following example.
curl -X POST -H "Content-Type: application/json" -d '{
"query": "{your-expensive-query}"
}' "http://{username}.cartodb.com/api/v2/sql/job?api_key={api_key}"
If you are using the Batch API create operation for a Node.js client POST request with private tables, use the following code:
var request = require("request");
var options = {
method: 'POST',
url: 'http://{username}.cartodb.com/api/v2/sql/job',
qs: {
'api_key': '{api_key}'
},
headers: {
'content-type': 'application/json'
},
body: {
query: '{your-expensive-query}'
},
json: true
};
request(options, function (error, response, body) {
if (error) throw new Error(error);
console.log(body);
});