1.5.4. /db/_design/design-doc/_view/view-name
Executes the specified view function from the specified design document.
Request:
Response:
HTTP/1.1 200 OK
Cache-Control: must-revalidate
Content-Type: application/json
Date: Wed, 21 Aug 2013 09:12:06 GMT
ETag: "2FOLSBSW4O6WB798XU4AQYA9B"
Server: CouchDB (Erlang/OTP)
Transfer-Encoding: chunked
{
"offset": 0,
"rows": [
{
"id": "SpaghettiWithMeatballs",
"key": "meatballs",
"value": 1
},
{
"id": "SpaghettiWithMeatballs",
"key": "spaghetti",
"value": 1
},
{
"id": "SpaghettiWithMeatballs",
"key": "tomato sauce",
"value": 1
}
],
"total_rows": 3
}
Changed in version 1.6.0: added attachments
and att_encoding_info
parameters
Changed in version 2.0.0: added sorted
parameter
Changed in version 2.1.0: added stable
and update
parameters
POST
/{db}/_design/{ddoc}/_view/{view}
Executes the specified view function from the specified design document. POST view functionality supports identical parameters and behavior as specified in the API but allows for the query string parameters to be supplied as keys in a JSON object in the body of the POST request.
Request:
POST /recipes/_design/ingredients/_view/by_name HTTP/1.1
Accept: application/json
Content-Length: 37
Host: localhost:5984
{
"keys": [
"meatballs",
"spaghetti"
]
}
Response:
HTTP/1.1 200 OK
Cache-Control: must-revalidate
Content-Type: application/json
Date: Wed, 21 Aug 2013 09:14:13 GMT
ETag: "6R5NM8E872JIJF796VF7WI3FZ"
Server: CouchDB (Erlang/OTP)
Transfer-Encoding: chunked
{
"offset": 0,
"rows": [
{
"id": "SpaghettiWithMeatballs",
"key": "meatballs",
"value": 1
},
{
"id": "SpaghettiWithMeatballs",
"key": "spaghetti",
"value": 1
}
],
"total_rows": 3
}
There are two view indexing options that can be defined in a design document as boolean properties of an options
object. Unlike the others querying options, these aren’t URL parameters because they take effect when the view index is generated, not when it’s accessed:
- local_seq (boolean): Makes documents’ local sequence numbers available to map functions (as a
_local_seq
document property) - include_design (boolean): Allows map functions to be called on design documents as well as regular documents
The definition of a view within a design document also creates an index based on the key information defined within each view. The production and use of the index significantly increases the speed of access and searching or selecting documents from the view.
However, the index is not updated when new documents are added or modified in the database. Instead, the index is generated or updated, either when the view is first accessed, or when the view is accessed after a document has been updated. In each case, the index is updated before the view query is executed against the database.
View indexes are updated incrementally in the following situations:
- A new document has been added to the database.
- A document has been deleted from the database.
- A document in the database has been updated.
Because the view is updated when it has been queried, it can result in a delay in returned information when the view is accessed, especially if there are a large number of documents in the database and the view index does not exist. There are a number of ways to mitigate, but not completely eliminate, these issues. These include:
- Create the view definition (and associated design documents) on your database before allowing insertion or updates to the documents. If this is allowed while the view is being accessed, the index can be updated incrementally.
- Manually force a view request from the database. You can do this either before users are allowed to use the view, or you can access the view manually after documents are added or updated.
- Use the changes feed to monitor for changes to the database and then access the view to force the corresponding view index to be updated.
None of these can completely eliminate the need for the indexes to be rebuilt or updated when the view is accessed, but they may lessen the effects on end-users of the index update affecting the user experience.
Another alternative is to allow users to access a ‘stale’ version of the view index, rather than forcing the index to be updated and displaying the updated results. Using a stale view may not return the latest information, but will return the results of the view query using an existing version of the index.
For example, to access the existing stale view by_recipe
in the recipes
design document:
http://localhost:5984/recipes/_design/recipes/_view/by_recipe?stale=ok
Accessing a stale view:
- Does not trigger a rebuild of the view indexes, even if there have been changes since the last access.
- Returns the current version of the view index, if a current version exists.
- Returns an empty result set if the given view index does not exist.
As an alternative, you use the update_after
value to the stale
parameter. This causes the view to be returned as a stale view, but for the update process to be triggered after the view information has been returned to the client.
In addition to using stale views, you can also make use of the update_seq
query argument. Using this query argument generates the view information including the update sequence of the database from which the view was generated. The returned value can be compared this to the current update sequence exposed in the database information (returned by ).
Each element within the returned array is sorted using native UTF-8 sorting according to the contents of the key portion of the emitted content. The basic order of output is as follows:
null
false
true
- Numbers
- Text (case sensitive, lowercase first)
- Arrays (according to the values of each element, in order)
- Objects (according to the values of keys, in key order)
Request:
GET /db/_design/test/_view/sorting HTTP/1.1
Accept: application/json
Host: localhost:5984
Response:
You can reverse the order of the returned view information by using the descending
query value set to true:
Request:
GET /db/_design/test/_view/sorting?descending=true HTTP/1.1
Accept: application/json
Host: localhost:5984
Response:
HTTP/1.1 200 OK
Cache-Control: must-revalidate
Content-Type: application/json
Date: Wed, 21 Aug 2013 10:09:25 GMT
ETag: "Z4N468R15JBT98OM0AMNSR8U"
Server: CouchDB (Erlang/OTP)
Transfer-Encoding: chunked
{
"rows": [
"id": "dummy-doc",
"key": {
"foo": "bar"
},
"value": null
},
{
"id": "dummy-doc",
"key": {},
"value": null
},
{
"id": "dummy-doc",
"key": [
3
],
"value": null
},
{
"id": "dummy-doc",
"key": [
2,
3
],
"value": null
},
{
"id": "dummy-doc",
"key": [
1,
2,
3
],
"value": null
},
{
"id": "dummy-doc",
"key": [],
"value": null
},
{
"id": "dummy-doc",
"key": "\u043f\u0440\u0438\u0432\u0435\u0442",
"value": null
},
{
"id": "dummy-doc",
"key": "Hello",
"value": null
},
{
"id": "dummy-doc",
"key": "hello",
"value": null
},
{
"id": "dummy-doc",
"key": "10",
"value": null
},
{
"id": "dummy-doc",
"key": 42,
"value": null
},
{
"id": "dummy-doc",
"key": 10,
"value": null
},
{
"id": "dummy-doc",
"key": 1,
"value": null
},
{
"id": "dummy-doc",
"key": 0,
"value": null
},
{
"id": "dummy-doc",
"key": true,
"value": null
},
{
"id": "dummy-doc",
"key": false,
"value": null
},
{
"id": "dummy-doc",
"key": null,
"value": null
}
],
"total_rows": 17
}
The sorting direction is applied before the filtering applied using the startkey
and endkey
query arguments. For example the following query:
GET http://couchdb:5984/recipes/_design/recipes/_view/by_ingredient?startkey=%22carrots%22&endkey=%22egg%22 HTTP/1.1
Accept: application/json
GET /recipes/_design/recipes/_view/by_ingredient?descending=true&startkey=%22carrots%22&endkey=%22egg%22 HTTP/1.1
Accept: application/json
Host: localhost:5984
{
"total_rows" : 26453,
"rows" : [],
"offset" : 21882
}
The results will be empty because the entries in the view are reversed before the key filter is applied, and therefore the endkey
of “egg” will be seen before the startkey
of “carrots”, resulting in an empty list.
Instead, you should reverse the values supplied to the startkey
and endkey
parameters to match the descending sorting applied to the keys. Changing the previous example to:
GET /recipes/_design/recipes/_view/by_ingredient?descending=true&startkey=%22egg%22&endkey=%22carrots%22 HTTP/1.1
Accept: application/json
Host: localhost:5984
1.5.4.3.2. Raw collation
By default CouchDB uses an driver for sorting view results. It’s possible use binary collation instead for faster view builds where Unicode collation is not important.
To use raw collation add "collation": "raw"
key-value pair to the design documents options
object at the root level. After that, views will be regenerated and new order applied.
By default, views return all results. That’s ok when the number of results is small, but this may lead to problems when there are billions results, since the client may have to read them all and consume all available memory.
But it’s possible to reduce output result rows by specifying limit
query parameter. For example, retrieving the list of recipes using the by_title
view and limited to 5 returns only 5 records, while there are total 2667 records in view:
Request:
Response:
HTTP/1.1 200 OK
Cache-Control: must-revalidate
Content-Type: application/json
Date: Wed, 21 Aug 2013 09:14:13 GMT
ETag: "9Q6Q2GZKPH8D5F8L7PB6DBSS9"
Server: CouchDB (Erlang/OTP)
Transfer-Encoding: chunked
{
"offset" : 0,
"rows" : [
{
"id" : "3-tiersalmonspinachandavocadoterrine",
"key" : "3-tier salmon, spinach and avocado terrine",
"value" : [
null,
"3-tier salmon, spinach and avocado terrine"
]
},
{
"id" : "Aberffrawcake",
"key" : "Aberffraw cake",
"value" : [
null,
"Aberffraw cake"
]
{
"id" : "Adukiandorangecasserole-microwave",
"key" : "Aduki and orange casserole - microwave",
"value" : [
null,
"Aduki and orange casserole - microwave"
]
},
{
"id" : "Aioli-garlicmayonnaise",
"key" : "Aioli - garlic mayonnaise",
"value" : [
null,
]
},
{
"id" : "Alabamapeanutchicken",
"key" : "Alabama peanut chicken",
"value" : [
null,
"Alabama peanut chicken"
]
}
],
"total_rows" : 2667
}
To omit some records you may use skip
query parameter:
Request:
GET /recipes/_design/recipes/_view/by_title?limit=3&skip=2 HTTP/1.1
Accept: application/json
Host: localhost:5984
Response:
HTTP/1.1 200 OK
Cache-Control: must-revalidate
Content-Type: application/json
Date: Wed, 21 Aug 2013 09:14:13 GMT
ETag: "H3G7YZSNIVRRHO5FXPE16NJHN"
Server: CouchDB (Erlang/OTP)
Transfer-Encoding: chunked
{
"offset" : 2,
"rows" : [
{
"id" : "Adukiandorangecasserole-microwave",
"key" : "Aduki and orange casserole - microwave",
"value" : [
null,
"Aduki and orange casserole - microwave"
]
},
{
"id" : "Aioli-garlicmayonnaise",
"key" : "Aioli - garlic mayonnaise",
"value" : [
null,
"Aioli - garlic mayonnaise"
]
},
{
"id" : "Alabamapeanutchicken",
"key" : "Alabama peanut chicken",
"value" : [
null,
"Alabama peanut chicken"
]
}
],
"total_rows" : 2667
}
New in version 2.2.
POST
/{db}/_design/{ddoc}/_view/{view}/queries
Executes multiple specified view queries against the view function from the specified design document.
Parameters: |
|
---|---|
Request Headers: | |
| |
Request JSON Object: | |
| |
Response Headers: | |
| |
Response JSON Object: | |
| |
Status Codes: |
|
POST /recipes/_design/recipes/_view/by_title/queries HTTP/1.1
Content-Type: application/json
Accept: application/json
Host: localhost:5984
{
"queries": [
{
"keys": [
"meatballs",
"spaghetti"
]
},
{
"limit": 3,
"skip": 2
}
]
}
Response:
HTTP/1.1 200 OK
Cache-Control: must-revalidate
Content-Type: application/json
Date: Wed, 20 Dec 2016 11:17:07 GMT
ETag: "1H8RGBCK3ABY6ACDM7ZSC30QK"
Server: CouchDB (Erlang/OTP)
Transfer-Encoding: chunked
{
"results" : [
{
"offset": 0,
"rows": [
{
"id": "SpaghettiWithMeatballs",
"key": "meatballs",
"value": 1
},
{
"id": "SpaghettiWithMeatballs",
"key": "spaghetti",
"value": 1
},
{
"id": "SpaghettiWithMeatballs",
"key": "tomato sauce",
"value": 1
}
],
"total_rows": 3
},
{
"offset" : 2,
"rows" : [
{
"id" : "Adukiandorangecasserole-microwave",
"key" : "Aduki and orange casserole - microwave",
"value" : [
null,
"Aduki and orange casserole - microwave"
]
},
{
"id" : "Aioli-garlicmayonnaise",
"key" : "Aioli - garlic mayonnaise",
"value" : [
null,
"Aioli - garlic mayonnaise"
]
},
{
"id" : "Alabamapeanutchicken",
"key" : "Alabama peanut chicken",
"value" : [
null,
"Alabama peanut chicken"
]
}
],
"total_rows" : 2667
}