couchdb-iterator
A fast and easy to ease CouchDB iterator for views and all documents.
Installation
$ npm install couchdb-iterator
Usage
Concurrent row iteration
couchdbIterator(couchdbAddr, [view], iterator, [options])
Calls iterator
for each row of the database referenced by couchdbAddr
.
If a view
is supplied, iterates only over that view's rows.
This library aims to be fast, therefore iteration happens concurrently. The iterator function can be async but beware that order
is not guaranteed.
Examples:
const couchdbIterator = require('couchdb-iterator');
couchdbIterator('http://localhost:5984/my-db', (row, index) => {
console.log(index, row.id, row.key, row.value);
})
.then((rowsCount) => {
console.log(`Iteration completed! ${rowsCount}`);
}, (err) => {
console.log('Iteration failed', err);
});
couchdbIterator('http://localhost:5984/my-db', 'my-design-doc/my-view', (row, index) => {
console.log(index, row.id, row.key, row.value);
})
.then((rowsCount) => {
console.log(`Iteration completed! ${rowsCount}`);
}, (err) => {
console.log('Iteration failed', err);
});
The couchdbAddr
argument must be a connection string with protocol, host, port and database path (e.g.: http://localhost:5984/my-db) or a nano instance. The view
argument is a string in the form of design-doc/view-name
(e.g.: app/byUser).
Available options:
concurrency
: The concurrency in which the iterator
is called, defaults to 50
.nano
: Custom options to be used when creating the nano instance, defaults to null
.- The following querying options are available:
limit
, skip
, stale
, descending
, startkey
, startkey_docid
, endkey
, endkey_docid
, include_docs
and inclusive_end
(can be camelCased).
All querying options have no default value, except for limit
which is concurrency * 10
. Also, stale
is automatically set to ok
after the first iteration to further improve performance.
Bulk iteration
couchdbIterator.bulk(couchdbAddr, [view], iterator, [options])
Calls iterator
for a bulk of rows of the database referenced by couchdbAddr
.
If a view
is supplied, iterates only over that view's rows.
This method is similar to couchdbIterator()
but iterates in bulks and it respects the order of rows. The order is respected because since a bulk is ordered and the next bulk only comes when the current bulk is handled.
Examples:
const couchdbIterator = require('couchdb-iterator');
couchdbIterator.bulk('http://localhost:5984/my-db', (rows) => {
rows.forEach((row) => {
console.log(row.index, row.id, row.key, row.value);
});
})
.then((rowsCount) => {
console.log(`Iteration completed! ${rowsCount}`);
}, (err) => {
console.log('Iteration failed', err);
});
couchdbIterator.bulk('http://localhost:5984/my-db', 'my-design-doc/my-view', (rows) => {
rows.forEach((row) => {
console.log(row.index, row.id, row.key, row.value);
});
})
.then((rowsCount) => {
console.log(`Iteration completed! ${rowsCount}`);
}, (err) => {
console.log('Iteration failed', err);
});
The couchdbAddr
argument must be a connection string with protocol, host, port and database path (e.g.: http://localhost:5984/my-db) or a nano instance. The view
argument is a string in the form of design-doc/view-name
(e.g.: app/byUser).
Available options:
bulkSize
: The bulkSize, defaults to 50
.nano
: Custom options to be used when creating the nano instance, defaults to null
.- The following querying options are available:
limit
, skip
, stale
, descending
, startkey
, startkey_docid
, endkey
, endkey_docid
, include_docs
and inclusive_end
(can be camelCased).
All querying options have no default value, except for limit
which is bulkSize * 10
. Also, stale
is automatically set to ok
after the first iteration to further improve performance.
Tests
$ npm test
$ npm test-cov
to get coverage report
The tests expect a running CouchDB in http://localhost:5984
and will create and destroy couchdb-iterator-tests
database. You may specify a different address with COUCHDB
, e.g.: $ COUCHDB=http://admin:admin@localhost:5984/my-custom-database-for-tests npm test
.
License
Released under the MIT License.