API:Opensearch
![]() | This page is part of the MediaWiki Action API documentation. |
GET request to search wiki pages by title, this uses the OpenSearch format.
API documentation
![]() | The following documentation is the output of Special: |
action=opensearch
- This module requires read rights.
- Source: MediaWiki
- License: GPL-2.0-or-later
Search the wiki using the OpenSearch protocol.
- search
Search string.
- This parameter is required.
- namespace
Namespaces to search. Ignored if search begins with a valid namespace prefix.
- Values (separate with | or alternative): 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 90, 91, 92, 93, 100, 101, 102, 103, 104, 105, 106, 107, 710, 711, 828, 829, 1198, 1199, 2600, 5500, 5501
- To specify all values, use *.
- Default: 0
- limit
Maximum number of results to return.
- Type: integer or max
- The value must be between 1 and 500.
- Default: 10
- profile
Search profile to use.
- strict
- Strict profile with few punctuation characters removed but diacritics and stress marks are kept.
- normal
- Few punctuation characters, some diacritics and stopwords removed.
- normal-subphrases
- Few punctuation characters, some diacritics and stopwords removed. It will match also subphrases (can be subphrases or subpages depending on internal wiki configuration).
- fuzzy
- Similar to normal with typo correction (two typos supported).
- fast-fuzzy
- Experimental fuzzy profile (may be removed at any time)
- fuzzy-subphrases
- Similar to normal with typo correction (two typos supported). It will match also subphrases (can be subphrases or subpages depending on internal wiki configuration).
- classic
- Classic prefix, few punctuation characters and some diacritics removed.
- engine_autoselect
- Let the search engine decide on the best profile to use.
- One of the following values: classic, engine_autoselect, fast-fuzzy, fuzzy, fuzzy-subphrases, normal, normal-subphrases, strict
- Default: engine_autoselect
- suggest
- Deprecated.
No longer used.
- Type: boolean (details)
- redirects
How to handle redirects:
- return
- Return the redirect itself.
- resolve
- Return the target page. May return fewer than limit results.
For historical reasons, the default is "return" for format=json and "resolve" for other formats.
- One of the following values: resolve, return
- format
The format of the output.
- One of the following values: json, jsonfm, xml, xmlfm
- Default: json
- warningsaserror
If warnings are raised with format=json, return an API error instead of ignoring them.
- Type: boolean (details)
- Find pages beginning with Te.
- api.php?action=opensearch&search=Te [open in sandbox]
Example
GET request
Response
[
"Hampi",
[
"Hampi",
"Hampi (town)",
"Hampi Express",
...
],
[
"Hampi, also referred to as the Group of Monuments at Hampi, is a UNESCO World Heritage Site located in east-central Karnataka, India.",
"Hampi is a town in Hospet taluk of the Ballari district in the Indian state of Karnataka. Located along the Tungabhadra River in the east and center part of the state, near the border of Andhra Pradesh, Hampi is near the city of Hosapete.",
"The Hampi Express is a daily express train running between the Mysooru and Hubballi Junction, the headquarters of the South Western Railway in India.",
...
],
[
"https://en.wikipedia.org/wiki/Hampi",
"https://en.wikipedia.org/wiki/Hampi_(town)",
"https://en.wikipedia.org/wiki/Hampi_Express",
...
]
]
Sample code
Python
#!/usr/bin/python3
"""
opensearch.py
MediaWiki API Demos
Demo of `Opensearch` module: Search the wiki and obtain
results in an OpenSearch (http://www.opensearch.org) format
MIT License
"""
import requests
S = requests.Session()
URL = "https://en.wikipedia.org/w/api.php"
PARAMS = {
"action": "opensearch",
"namespace": "0",
"search": "Hampi",
"limit": "5",
"format": "json"
}
R = S.get(url=URL, params=PARAMS)
DATA = R.json()
print(DATA)
PHP
<?php
/*
opensearch.php
MediaWiki API Demos
Demo of `Opensearch` module: Search the wiki and obtain
results in an OpenSearch (http://www.opensearch.org) format
MIT License
*/
$endPoint = "https://en.wikipedia.org/w/api.php";
$params = [
"action" => "opensearch",
"search" => "Hampi",
"limit" => "5",
"namespace" => "0",
"format" => "json"
];
$url = $endPoint . "?" . http_build_query( $params );
$ch = curl_init( $url );
curl_setopt( $ch, CURLOPT_RETURNTRANSFER, true );
$output = curl_exec( $ch );
curl_close( $ch );
$result = json_decode( $output, true );
var_dump( $result );
JavaScript
/*
opensearch.js
MediaWiki API Demos
Demo of `Opensearch` module: Search the wiki and obtain
results in an OpenSearch (http://www.opensearch.org) format
MIT License
*/
var url = "https://en.wikipedia.org/w/api.php";
var params = {
action: "opensearch",
search: "Hampi",
limit: "5",
namespace: "0",
format: "json"
};
url = url + "?origin=*";
Object.keys(params).forEach(function(key){url += "&" + key + "=" + params[key];});
fetch(url)
.then(function(response){return response.json();})
.then(function(response) {console.log(response);})
.catch(function(error){console.log(error);});
MediaWiki JS
/*
opensearch.js
MediaWiki API Demos
Demo of `Opensearch` module: Search the wiki and obtain
results in an OpenSearch (http://www.opensearch.org) format
MIT License
*/
var params = {
action: 'opensearch',
search: 'Hampi',
limit: '5',
namespace: '0',
format: 'json'
},
api = new mw.Api();
api.get( params ).done( function ( data ) {
console.log( data );
} );
Possible errors
Code | Info |
---|---|
nosearch | The search parameter must be set. |
unknown_format | Unrecognized value for parameter format: aaa. |
Additional notes
Related to this API, a few additional pointers for MediaWiki site administrators and extension developers:
- Extension:TitleKey - allows search suggestions from this API to be case insensitive.
- Extension:TextExtracts and
$wgExtractsExtendOpenSearchXml
configured as true inLocalSettings.php
would allow each Item in the XML format to include a<Description>
tag with a text extract from the article.
- Extension:PageImages and
$wgPageImagesExpandOpenSearchXml
configured astrue
inLocalSettings.php
would allow each item in the XML format to include an<Image>
tag with a thumbnail image from the article.